https://umesh-malik.com/blog/eliminate-pcie-bottleneck-ai-training 2026-08-25 daily 1.0 https://umesh-malik.com/blog/eliminate-pcie-bottleneck-ai-training-cover.png Cover showing the PCIe bottleneck in traditional GPU-NIC architecture versus MTIA 300's built-in NIC design that delivers 1.2 TB/s without CPU mediation Cover showing the PCIe bottleneck in traditional GPU-NIC architecture versus MTIA 300's built-in NIC design that delivers 1.2 TB/s without CPU mediation https://umesh-malik.com/blog/eliminate-pcie-bottleneck-ai-training-pcie-path.png Cover showing the PCIe bottleneck in traditional GPU-NIC architecture versus MTIA 300's built-in NIC design that delivers 1.2 TB/s without CPU mediation Cover showing the PCIe bottleneck in traditional GPU-NIC architecture versus MTIA 300's built-in NIC design that delivers 1.2 TB/s without CPU mediation https://umesh-malik.com/blog/eliminate-pcie-bottleneck-ai-training-mtia-arch.png Cover showing the PCIe bottleneck in traditional GPU-NIC architecture versus MTIA 300's built-in NIC design that delivers 1.2 TB/s without CPU mediation Cover showing the PCIe bottleneck in traditional GPU-NIC architecture versus MTIA 300's built-in NIC design that delivers 1.2 TB/s without CPU mediation Umesh Malik en 2026-08-25 Fix the PCIe Bottleneck in AI Training: How Built-in NICs Work https://umesh-malik.com/blog/ms-paint-invisible-watermark-guid 2026-08-25 daily 1.0 https://umesh-malik.com/blog/ms-paint-invisible-watermark-guid-cover.png The flow from prompt to invisible watermark: Paint sends a prompt to Microsoft's moderation server, receives a GUID, runs local inference, then embeds that GUID into the pixels The flow from prompt to invisible watermark: Paint sends a prompt to Microsoft's moderation server, receives a GUID, runs local inference, then embeds that GUID into the pixels https://umesh-malik.com/blog/ms-paint-invisible-watermark-guid-flow.png The flow from prompt to invisible watermark: Paint sends a prompt to Microsoft's moderation server, receives a GUID, runs local inference, then embeds that GUID into the pixels The flow from prompt to invisible watermark: Paint sends a prompt to Microsoft's moderation server, receives a GUID, runs local inference, then embeds that GUID into the pixels https://umesh-malik.com/blog/ms-paint-invisible-watermark-guid-local-vs-remote.png The flow from prompt to invisible watermark: Paint sends a prompt to Microsoft's moderation server, receives a GUID, runs local inference, then embeds that GUID into the pixels The flow from prompt to invisible watermark: Paint sends a prompt to Microsoft's moderation server, receives a GUID, runs local inference, then embeds that GUID into the pixels Umesh Malik en 2026-08-25 MS Paint Invisible Watermark: How to Find the GUID in AI Images https://umesh-malik.com/blog/secure-llm-inference-vllm-cve-2025-9141 2026-08-25 daily 1.0 https://umesh-malik.com/blog/secure-llm-inference-vllm-cve-2025-9141-cover.png Cover showing the inference engine attack surface with token stream flowing from model through vulnerable parser to arbitrary code execution, and the defense architecture separating GPU host from token parsing Cover showing the inference engine attack surface with token stream flowing from model through vulnerable parser to arbitrary code execution, and the defense architecture separating GPU host from token parsing https://umesh-malik.com/blog/secure-llm-inference-vllm-cve-2025-9141-attack-surface.png Cover showing the inference engine attack surface with token stream flowing from model through vulnerable parser to arbitrary code execution, and the defense architecture separating GPU host from token parsing Cover showing the inference engine attack surface with token stream flowing from model through vulnerable parser to arbitrary code execution, and the defense architecture separating GPU host from token parsing https://umesh-malik.com/blog/secure-llm-inference-vllm-cve-2025-9141-defense-architecture.png Cover showing the inference engine attack surface with token stream flowing from model through vulnerable parser to arbitrary code execution, and the defense architecture separating GPU host from token parsing Cover showing the inference engine attack surface with token stream flowing from model through vulnerable parser to arbitrary code execution, and the defense architecture separating GPU host from token parsing Umesh Malik en 2026-08-25 How to Harden vLLM Inference: CVE-2025-9141 Defense Guide https://umesh-malik.com/blog/chatgpt-search-site-scoping-geo 2026-08-24 monthly 0.8 https://umesh-malik.com/blog/chatgpt-search-site-scoping-geo-cover.png Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 https://umesh-malik.com/blog/chatgpt-search-site-scoping-geo-data.png Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 https://umesh-malik.com/blog/chatgpt-search-site-scoping-geo-flow.png Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 https://umesh-malik.com/blog/chatgpt-search-site-scoping-geo-factors.png Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 Chart showing ChatGPT Search site-scoped query share jumping from 0.3% to 17% on August 8, 2026 https://umesh-malik.com/blog/sync-robots-txt-ai-bot-blocks 2026-08-24 monthly 0.8 https://umesh-malik.com/blog/sync-robots-txt-ai-bot-blocks-cover.png Diagram showing robots.txt preferences syncing with Cloudflare's AI bot enforcement rules across Search, Agent, and Training categories Diagram showing robots.txt preferences syncing with Cloudflare's AI bot enforcement rules across Search, Agent, and Training categories https://umesh-malik.com/blog/sync-robots-txt-ai-bot-blocks-flow.png Diagram showing robots.txt preferences syncing with Cloudflare's AI bot enforcement rules across Search, Agent, and Training categories Diagram showing robots.txt preferences syncing with Cloudflare's AI bot enforcement rules across Search, Agent, and Training categories https://umesh-malik.com/blog/sync-robots-txt-ai-bot-blocks-decision.png Diagram showing robots.txt preferences syncing with Cloudflare's AI bot enforcement rules across Search, Agent, and Training categories Diagram showing robots.txt preferences syncing with Cloudflare's AI bot enforcement rules across Search, Agent, and Training categories https://umesh-malik.com/blog/optional-oauth-scopes-mcp-servers 2026-08-23 monthly 0.8 https://umesh-malik.com/blog/optional-oauth-scopes-mcp-servers-cover.png OAuth consent screen with optional scope checkboxes that let users narrow agent permissions OAuth consent screen with optional scope checkboxes that let users narrow agent permissions https://umesh-malik.com/blog/optional-oauth-scopes-mcp-servers-consent-flow.png OAuth consent screen with optional scope checkboxes that let users narrow agent permissions OAuth consent screen with optional scope checkboxes that let users narrow agent permissions https://umesh-malik.com/blog/optional-oauth-scopes-mcp-servers-scope-evaluation.png OAuth consent screen with optional scope checkboxes that let users narrow agent permissions OAuth consent screen with optional scope checkboxes that let users narrow agent permissions https://umesh-malik.com/blog/optional-oauth-scopes-mcp-servers-token-flow.png OAuth consent screen with optional scope checkboxes that let users narrow agent permissions OAuth consent screen with optional scope checkboxes that let users narrow agent permissions https://umesh-malik.com/blog/rust-glancer-low-memory-lsp 2026-08-23 monthly 0.8 https://umesh-malik.com/blog/rust-glancer-low-memory-lsp-cover.png Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB https://umesh-malik.com/blog/rust-glancer-low-memory-lsp-architecture.png Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB https://umesh-malik.com/blog/rust-glancer-low-memory-lsp-benchmarks.png Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB https://umesh-malik.com/blog/rust-glancer-low-memory-lsp-dev-loop.png Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB Memory comparison between rust-analyzer and Rust Glancer showing the dramatic reduction from gigabytes to under 100MB https://umesh-malik.com/blog/self-hosted-ai-coding-agent-sandbox 2026-08-23 monthly 0.8 https://umesh-malik.com/blog/self-hosted-ai-coding-agent-sandbox-cover.png Architecture diagram of a self-hosted AI coding agent with sandboxed execution Architecture diagram of a self-hosted AI coding agent with sandboxed execution https://umesh-malik.com/blog/self-hosted-ai-coding-agent-sandbox-architecture.png Architecture diagram of a self-hosted AI coding agent with sandboxed execution Architecture diagram of a self-hosted AI coding agent with sandboxed execution https://umesh-malik.com/blog/self-hosted-ai-coding-agent-sandbox-cost.png Architecture diagram of a self-hosted AI coding agent with sandboxed execution Architecture diagram of a self-hosted AI coding agent with sandboxed execution https://umesh-malik.com/blog/self-hosted-ai-coding-agent-sandbox-risk.png Architecture diagram of a self-hosted AI coding agent with sandboxed execution Architecture diagram of a self-hosted AI coding agent with sandboxed execution https://umesh-malik.com/blog/vercel-cli-dns-commands 2026-08-23 monthly 0.8 https://umesh-malik.com/blog/vercel-cli-dns-commands-cover.png Vercel CLI DNS management workflow showing terminal commands for DNS records, domains, and project control Vercel CLI DNS management workflow showing terminal commands for DNS records, domains, and project control https://umesh-malik.com/blog/vercel-cli-dns-commands-update-flow.png Vercel CLI DNS management workflow showing terminal commands for DNS records, domains, and project control Vercel CLI DNS management workflow showing terminal commands for DNS records, domains, and project control https://umesh-malik.com/blog/vercel-cli-dns-commands-project-lifecycle.png Vercel CLI DNS management workflow showing terminal commands for DNS records, domains, and project control Vercel CLI DNS management workflow showing terminal commands for DNS records, domains, and project control https://umesh-malik.com/blog/claude-code-macos-printer-driver 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/claude-code-macos-printer-driver-cover.png The four-step path from HP's unsupported printer to a native macOS driver: Docker workaround, byte-diff against an open-source filter, the DPI bug isolated, then a native USB backend The four-step path from HP's unsupported printer to a native macOS driver: Docker workaround, byte-diff against an open-source filter, the DPI bug isolated, then a native USB backend https://umesh-malik.com/blog/claude-code-macos-printer-driver-dpi-bug.png The four-step path from HP's unsupported printer to a native macOS driver: Docker workaround, byte-diff against an open-source filter, the DPI bug isolated, then a native USB backend The four-step path from HP's unsupported printer to a native macOS driver: Docker workaround, byte-diff against an open-source filter, the DPI bug isolated, then a native USB backend https://umesh-malik.com/blog/claude-code-macos-printer-driver-architecture.png The four-step path from HP's unsupported printer to a native macOS driver: Docker workaround, byte-diff against an open-source filter, the DPI bug isolated, then a native USB backend The four-step path from HP's unsupported printer to a native macOS driver: Docker workaround, byte-diff against an open-source filter, the DPI bug isolated, then a native USB backend https://umesh-malik.com/blog/gemini-agent-hooks-fail-open 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/gemini-agent-hooks-fail-open-cover.png The five sandbox tools a Gemini agent hook intercepts, against the two tool families it never sees The five sandbox tools a Gemini agent hook intercepts, against the two tool families it never sees https://umesh-malik.com/blog/gemini-agent-hooks-fail-open-paths.png The five sandbox tools a Gemini agent hook intercepts, against the two tool families it never sees The five sandbox tools a Gemini agent hook intercepts, against the two tool families it never sees https://umesh-malik.com/blog/gemini-agent-hooks-fail-open-coverage.png The five sandbox tools a Gemini agent hook intercepts, against the two tool families it never sees The five sandbox tools a Gemini agent hook intercepts, against the two tool families it never sees https://umesh-malik.com/blog/rust-safe-gpu-offload-benchmarks 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/rust-safe-gpu-offload-benchmarks-cover.png Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP https://umesh-malik.com/blog/rust-safe-gpu-offload-benchmarks-pipeline.png Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP https://umesh-malik.com/blog/rust-safe-gpu-offload-benchmarks-comparison.png Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP https://umesh-malik.com/blog/rust-safe-gpu-offload-benchmarks-transfer-bug.png Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP Rust GPU offload safety and performance: ownership rules managing host-device transfers, benchmarked against hand-tuned CUDA and HIP https://umesh-malik.com/blog/svg-to-mp4-in-the-browser 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/svg-to-mp4-in-the-browser-cover.png A browser tab converting an animated SVG into an MP4 video entirely client-side using a 30MB ffmpeg.wasm build, with no server round trip A browser tab converting an animated SVG into an MP4 video entirely client-side using a 30MB ffmpeg.wasm build, with no server round trip https://umesh-malik.com/blog/svg-to-mp4-in-the-browser-workflow-comparison.png A browser tab converting an animated SVG into an MP4 video entirely client-side using a 30MB ffmpeg.wasm build, with no server round trip A browser tab converting an animated SVG into an MP4 video entirely client-side using a 30MB ffmpeg.wasm build, with no server round trip https://umesh-malik.com/blog/svg-to-mp4-in-the-browser-pipeline.png A browser tab converting an animated SVG into an MP4 video entirely client-side using a 30MB ffmpeg.wasm build, with no server round trip A browser tab converting an animated SVG into an MP4 video entirely client-side using a 30MB ffmpeg.wasm build, with no server round trip https://umesh-malik.com/blog/traffic-anomaly-or-outage-baseline-method 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/traffic-anomaly-or-outage-baseline-method-cover.png Cover graphic showing the four-step baseline method — 5-minute buckets, 3-week median, independent-cause match, verdict — next to the eclipse's 15-30% traffic-drop metrics Cover graphic showing the four-step baseline method — 5-minute buckets, 3-week median, independent-cause match, verdict — next to the eclipse's 15-30% traffic-drop metrics https://umesh-malik.com/blog/traffic-anomaly-or-outage-baseline-method-methodology.png Cover graphic showing the four-step baseline method — 5-minute buckets, 3-week median, independent-cause match, verdict — next to the eclipse's 15-30% traffic-drop metrics Cover graphic showing the four-step baseline method — 5-minute buckets, 3-week median, independent-cause match, verdict — next to the eclipse's 15-30% traffic-drop metrics https://umesh-malik.com/blog/traffic-anomaly-or-outage-baseline-method-baseline-comparison.png Cover graphic showing the four-step baseline method — 5-minute buckets, 3-week median, independent-cause match, verdict — next to the eclipse's 15-30% traffic-drop metrics Cover graphic showing the four-step baseline method — 5-minute buckets, 3-week median, independent-cause match, verdict — next to the eclipse's 15-30% traffic-drop metrics https://umesh-malik.com/blog/use-postgres-for-everything 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/use-postgres-for-everything-cover.png A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working https://umesh-malik.com/blog/use-postgres-for-everything-swaps.png A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working https://umesh-malik.com/blog/use-postgres-for-everything-vacuum-bill.png A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working https://umesh-malik.com/blog/use-postgres-for-everything-800jobs.png A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working A map of five infrastructure components replaced by PostgreSQL features, each annotated with the threshold where the swap stops working https://umesh-malik.com/blog/verify-ai-agent-benchmark-claims 2026-08-19 monthly 0.8 https://umesh-malik.com/blog/verify-ai-agent-benchmark-claims-cover.png A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating https://umesh-malik.com/blog/verify-ai-agent-benchmark-claims-collapse.png A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating https://umesh-malik.com/blog/verify-ai-agent-benchmark-claims-cost-collapse.png A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating https://umesh-malik.com/blog/verify-ai-agent-benchmark-claims-holdout-gate.png A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating A claimed 1.4x speedup collapsing to 1.5x slower as the same code is checked against a holdout benchmark and then audited for interface cheating https://umesh-malik.com/blog/turn-off-ai-features-permanently 2026-08-18 monthly 0.8 https://umesh-malik.com/blog/turn-off-ai-features-permanently-cover.png The four layers at which an AI feature can be switched off, ranked by which events reset each one The four layers at which an AI feature can be switched off, ranked by which events reset each one https://umesh-malik.com/blog/turn-off-ai-features-permanently-chrome-policy.png The four layers at which an AI feature can be switched off, ranked by which events reset each one The four layers at which an AI feature can be switched off, ranked by which events reset each one https://umesh-malik.com/blog/turn-off-ai-features-permanently-durability.png The four layers at which an AI feature can be switched off, ranked by which events reset each one The four layers at which an AI feature can be switched off, ranked by which events reset each one https://umesh-malik.com/blog/ai-writing-policy-for-engineers 2026-08-17 monthly 0.8 https://umesh-malik.com/blog/ai-writing-policy-for-engineers-cover.png An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop https://umesh-malik.com/blog/ai-writing-policy-for-engineers-lossy-chain.png An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop https://umesh-malik.com/blog/ai-writing-policy-for-engineers-four-rules.png An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop https://umesh-malik.com/blog/ai-writing-policy-for-engineers-effort-crossover.png An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop An author's intent passing through a draft and an AI rewrite before reaching the reader, losing emphasis, certainty and priority at each hop https://umesh-malik.com/blog/remove-cloudflare-beacon-min-js 2026-08-17 monthly 0.8 https://umesh-malik.com/blog/remove-cloudflare-beacon-min-js-cover.png Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser https://umesh-malik.com/blog/remove-cloudflare-beacon-min-js-injection-path.png Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser https://umesh-malik.com/blog/remove-cloudflare-beacon-min-js-optout-path.png Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser https://umesh-malik.com/blog/remove-cloudflare-beacon-min-js-csp-destinations.png Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser Cloudflare's edge injecting the beacon.min.js analytics script into an HTML response before it reaches the browser https://umesh-malik.com/blog/fix-eevdf-latency-regression-sched-ext 2026-08-16 monthly 0.8 https://umesh-malik.com/blog/fix-eevdf-latency-regression-sched-ext-cover.png Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF https://umesh-malik.com/blog/fix-eevdf-latency-regression-sched-ext-results.png Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF https://umesh-malik.com/blog/fix-eevdf-latency-regression-sched-ext-pools.png Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF https://umesh-malik.com/blog/fix-eevdf-latency-regression-sched-ext-safety.png Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF Two CPU pools under a sched_ext policy: latency-critical request threads separated from background work, with a watchdog fallback to EEVDF https://umesh-malik.com/blog/how-dna-wires-a-brain 2026-08-16 monthly 0.8 https://umesh-malik.com/blog/how-dna-wires-a-brain-cover.png Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits https://umesh-malik.com/blog/how-dna-wires-a-brain-budget-gap.png Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits https://umesh-malik.com/blog/how-dna-wires-a-brain-address-tree.png Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits https://umesh-malik.com/blog/how-dna-wires-a-brain-strategy-cost.png Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits Three brain-wiring strategies compared by information cost, from an infeasible 3x10^15 bits down to a genome-feasible 300 million bits https://umesh-malik.com/blog/zero-downtime-database-migration-dual-writes 2026-08-16 monthly 0.8 https://umesh-malik.com/blog/zero-downtime-database-migration-dual-writes-cover.png The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover https://umesh-malik.com/blog/zero-downtime-database-migration-dual-writes-ladder.png The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover https://umesh-malik.com/blog/zero-downtime-database-migration-dual-writes-percentile-gap.png The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover https://umesh-malik.com/blog/zero-downtime-database-migration-dual-writes-shadow-read-path.png The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover The five feature-flagged phases of a zero-downtime datastore migration, from dual writes through shadow reads to cutover https://umesh-malik.com/blog/cloudflare-access-for-workers 2026-08-15 monthly 0.8 https://umesh-malik.com/blog/cloudflare-access-for-workers-cover.png Cloudflare Access enforcing an identity policy in front of a Worker before its code executes Cloudflare Access enforcing an identity policy in front of a Worker before its code executes https://umesh-malik.com/blog/cloudflare-access-for-workers-request-path.png Cloudflare Access enforcing an identity policy in front of a Worker before its code executes Cloudflare Access enforcing an identity policy in front of a Worker before its code executes https://umesh-malik.com/blog/cloudflare-access-for-workers-entry-points.png Cloudflare Access enforcing an identity policy in front of a Worker before its code executes Cloudflare Access enforcing an identity policy in front of a Worker before its code executes https://umesh-malik.com/blog/cloudflare-access-for-workers-scopes.png Cloudflare Access enforcing an identity policy in front of a Worker before its code executes Cloudflare Access enforcing an identity policy in front of a Worker before its code executes https://umesh-malik.com/blog/qwen3-8-27b-vram-kv-cache-math 2026-08-15 monthly 0.8 https://umesh-malik.com/blog/qwen3-8-27b-vram-kv-cache-math-cover.png Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU https://umesh-malik.com/blog/qwen3-8-27b-vram-kv-cache-math-layers.png Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU https://umesh-malik.com/blog/qwen3-8-27b-vram-kv-cache-math-kv-growth.png Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU https://umesh-malik.com/blog/qwen3-8-27b-vram-kv-cache-math-budget.png Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU Qwen3.8 27B VRAM budget: FP8 weights plus KV cache at 262K context on a single GPU https://umesh-malik.com/blog/cut-agent-tool-call-cost-prompt-rewrite 2026-08-14 monthly 0.8 https://umesh-malik.com/blog/cut-agent-tool-call-cost-prompt-rewrite-cover.png Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored https://umesh-malik.com/blog/cut-agent-tool-call-cost-prompt-rewrite-rebill.png Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored https://umesh-malik.com/blog/cut-agent-tool-call-cost-prompt-rewrite-narrow-first.png Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored https://umesh-malik.com/blog/cut-agent-tool-call-cost-prompt-rewrite-trace.png Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored Swapping bespoke search tools for generic grep, glob and view raised Copilot code review cost until the system prompt was rewritten to be diff-anchored https://umesh-malik.com/blog/kubernetes-on-bare-metal-cloud-integrations 2026-08-14 monthly 0.8 https://umesh-malik.com/blog/kubernetes-on-bare-metal-cloud-integrations-cover.png The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal https://umesh-malik.com/blog/kubernetes-on-bare-metal-cloud-integrations-seams.png The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal https://umesh-malik.com/blog/kubernetes-on-bare-metal-cloud-integrations-lb-path.png The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal https://umesh-malik.com/blog/kubernetes-on-bare-metal-cloud-integrations-csi-blocker.png The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal The four integration seams a cloud provider fills for Kubernetes, and what you must build yourself on bare metal https://umesh-malik.com/blog/research-spike-as-running-code 2026-08-13 monthly 0.8 https://umesh-malik.com/blog/research-spike-as-running-code-cover.png A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed https://umesh-malik.com/blog/research-spike-as-running-code-blocker-map.png A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed https://umesh-malik.com/blog/research-spike-as-running-code-portability-line.png A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed https://umesh-malik.com/blog/research-spike-as-running-code-scope-ladder.png A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed A research spike shipped as running code surfaces engine-specific blockers a design doc would have missed https://umesh-malik.com/blog/verify-ai-crawler-ips-not-user-agents 2026-08-13 monthly 0.8 https://umesh-malik.com/blog/verify-ai-crawler-ips-not-user-agents-cover.png Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header https://umesh-malik.com/blog/verify-ai-crawler-ips-not-user-agents-freshness.png Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header https://umesh-malik.com/blog/verify-ai-crawler-ips-not-user-agents-failure-modes.png Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header https://umesh-malik.com/blog/verify-ai-crawler-ips-not-user-agents-ladder.png Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header Verifying AI crawler identity by matching source IPs against published CIDR lists instead of trusting the User-Agent header https://umesh-malik.com/blog/agent-context-compaction-what-survives 2026-08-12 monthly 0.8 https://umesh-malik.com/blog/agent-context-compaction-what-survives-cover.png How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped https://umesh-malik.com/blog/agent-context-compaction-what-survives-cutoff.png How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped https://umesh-malik.com/blog/agent-context-compaction-what-survives-billing.png How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped https://umesh-malik.com/blog/agent-context-compaction-what-survives-three-tools.png How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped How server-side compaction replaces a long agent transcript with a single summary block once input tokens cross the trigger, and what is kept versus permanently dropped https://umesh-malik.com/blog/on-device-ai-without-breaking-e2ee 2026-08-12 monthly 0.8 https://umesh-malik.com/blog/on-device-ai-without-breaking-e2ee-cover.png The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta https://umesh-malik.com/blog/on-device-ai-without-breaking-e2ee-telemetry-path.png The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta https://umesh-malik.com/blog/on-device-ai-without-breaking-e2ee-attestation-loop.png The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta https://umesh-malik.com/blog/on-device-ai-without-breaking-e2ee-naive-vs-verifiable.png The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta The metrics path behind WhatsApp Scam Alert, where device identity, IP address and per-user counters are each stripped at a different hop before any number reaches Meta https://umesh-malik.com/blog/fix-slow-llm-inference-macos-vms 2026-08-11 monthly 0.8 https://umesh-malik.com/blog/fix-slow-llm-inference-macos-vms-cover.png How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 https://umesh-malik.com/blog/fix-slow-llm-inference-macos-vms-capability-gate.png How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 https://umesh-malik.com/blog/fix-slow-llm-inference-macos-vms-throughput.png How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 https://umesh-malik.com/blog/fix-slow-llm-inference-macos-vms-recovery.png How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 How a paravirtualized Metal device reporting Apple GPU family 5 makes llama.cpp disable its simdgroup matrix kernels, and the throughput that returns when the guest reports family 9 https://umesh-malik.com/blog/learn-complex-topics-claude-code-simulations 2026-08-11 monthly 0.8 https://umesh-malik.com/blog/learn-complex-topics-claude-code-simulations-cover.png The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it https://umesh-malik.com/blog/learn-complex-topics-claude-code-simulations-loop.png The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it https://umesh-malik.com/blog/learn-complex-topics-claude-code-simulations-fit.png The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it https://umesh-malik.com/blog/learn-complex-topics-claude-code-simulations-tcp.png The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it The four-step loop for learning a complex topic by having an AI coding agent build an interactive simulation of it https://umesh-malik.com/blog/run-muse-glimmer-30b-locally 2026-08-11 monthly 0.8 https://umesh-malik.com/blog/run-muse-glimmer-30b-locally-cover.png Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression https://umesh-malik.com/blog/run-muse-glimmer-30b-locally-vram-ladder.png Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression https://umesh-malik.com/blog/run-muse-glimmer-30b-locally-dflash-speedup.png Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression https://umesh-malik.com/blog/run-muse-glimmer-30b-locally-runtime-matrix.png Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression Muse Glimmer 30B memory ladder showing 55GB at full precision shrinking to under 20GB with 4-bit K-Quant compression https://umesh-malik.com/blog/testing-llm-knowledge-cutoffs-opus-5 2026-08-11 monthly 0.8 https://umesh-malik.com/blog/testing-llm-knowledge-cutoffs-opus-5-cover.png Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal https://umesh-malik.com/blog/testing-llm-knowledge-cutoffs-opus-5-methodology.png Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal https://umesh-malik.com/blog/testing-llm-knowledge-cutoffs-opus-5-timeline.png Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal https://umesh-malik.com/blog/testing-llm-knowledge-cutoffs-opus-5-mimicry.png Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal Timeline comparing the knowledge-cutoff date vendors claim for Claude Opus 5 against the earlier cutoff its answers actually reveal https://umesh-malik.com/blog/sandbox-ai-agent-internet-access 2026-08-08 monthly 0.8 https://umesh-malik.com/blog/sandbox-ai-agent-internet-access-cover.png Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour https://umesh-malik.com/blog/sandbox-ai-agent-internet-access-attack-chain.png Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour https://umesh-malik.com/blog/sandbox-ai-agent-internet-access-numbers.png Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour https://umesh-malik.com/blog/sandbox-ai-agent-internet-access-controls.png Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour Cover showing the four-layer agent containment stack — deny egress, scope identity, watch live, kill fast — alongside the AISI evaluation figures of 122 runs, 10 off-scope runs and containment in under an hour https://umesh-malik.com/blog/vllm-throughput-tuning-flags 2026-08-08 monthly 0.8 https://umesh-malik.com/blog/vllm-throughput-tuning-flags-cover.png Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences https://umesh-malik.com/blog/vllm-throughput-tuning-flags-kv-budget.png Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences https://umesh-malik.com/blog/vllm-throughput-tuning-flags-scheduler-step.png Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences https://umesh-malik.com/blog/vllm-throughput-tuning-flags-symptom-map.png Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences Cover showing the vLLM VRAM budget split into model weights, runtime overhead and KV cache, with the KV cache block math that converts free VRAM into concurrent sequences https://umesh-malik.com/blog/ai-agent-permissions-approval-fatigue 2026-08-07 monthly 0.8 https://umesh-malik.com/blog/ai-agent-permissions-approval-fatigue-cover.png Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration https://umesh-malik.com/blog/ai-agent-permissions-approval-fatigue-miss-rates.png Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration https://umesh-malik.com/blog/ai-agent-permissions-approval-fatigue-npm-blindspot.png Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration https://umesh-malik.com/blog/ai-agent-permissions-approval-fatigue-layers.png Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration Cover card titled Approve, subtitled One in three gets through, with three statistics: 409,000 approve or deny decisions analysed, 66.3% mean accuracy, and 52.5% miss rate on npm-disguised exfiltration https://umesh-malik.com/blog/ai-gateway-for-workers-ai 2026-08-07 monthly 0.8 https://umesh-malik.com/blog/ai-gateway-for-workers-ai-cover.png Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop https://umesh-malik.com/blog/ai-gateway-for-workers-ai-path.png Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop https://umesh-malik.com/blog/ai-gateway-for-workers-ai-cache.png Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop https://umesh-malik.com/blog/ai-gateway-for-workers-ai-control-plane.png Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop Diagram of a Workers AI request routed through AI Gateway, showing the payload log, token count and cost attribution captured at the gateway hop https://umesh-malik.com/blog/datasette-sql-injection-patch 2026-08-07 monthly 0.8 https://umesh-malik.com/blog/datasette-sql-injection-patch-cover.png Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table https://umesh-malik.com/blog/datasette-sql-injection-patch-exposure.png Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table https://umesh-malik.com/blog/datasette-sql-injection-patch-path.png Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table https://umesh-malik.com/blog/datasette-sql-injection-patch-boundary.png Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table Diagram of the Datasette filter injection — a query-string column name flows unquoted into generated SQL and returns rows from a private table https://umesh-malik.com/blog/make-your-site-agent-readable 2026-08-07 monthly 0.8 https://umesh-malik.com/blog/make-your-site-agent-readable-cover.png The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site https://umesh-malik.com/blog/make-your-site-agent-readable-payload.png The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site https://umesh-malik.com/blog/make-your-site-agent-readable-layers.png The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site https://umesh-malik.com/blog/make-your-site-agent-readable-routing.png The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site The four agentic web layers — readable, discoverable, callable, payable — mapped to concrete URLs on a static site https://umesh-malik.com/blog/writing-agent-tool-instructions 2026-08-07 monthly 0.8 https://umesh-malik.com/blog/writing-agent-tool-instructions-cover.png The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce https://umesh-malik.com/blog/writing-agent-tool-instructions-browsing-loop.png The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce https://umesh-malik.com/blog/writing-agent-tool-instructions-three-tools.png The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce https://umesh-malik.com/blog/writing-agent-tool-instructions-cost.png The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce The browsing loop a code-review agent enters under generic tool instructions, versus the diff-anchored evidence path review-shaped instructions produce https://umesh-malik.com/blog/ads-for-ai-agents-time-markdown-crawlers 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/ads-for-ai-agents-time-markdown-crawlers-cover.png Ads for AI agents: TIME serves markdown with sponsored FAQ blocks to AI crawlers and HTML to humans Ads for AI agents: TIME serves markdown with sponsored FAQ blocks to AI crawlers and HTML to humans https://umesh-malik.com/blog/dependabot-grouped-updates-cut-pr-noise 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/dependabot-grouped-updates-cut-pr-noise-cover.png Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately https://umesh-malik.com/blog/dependabot-grouped-updates-cut-pr-noise-two-lanes.png Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately https://umesh-malik.com/blog/dependabot-grouped-updates-cut-pr-noise-volume.png Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately https://umesh-malik.com/blog/dependabot-grouped-updates-cut-pr-noise-key-map.png Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately Diagram of Dependabot running two independent lanes: version updates funnelled through cooldown and a monthly schedule into one grouped pull request, and security updates bypassing all of it to open immediately https://umesh-malik.com/blog/parallel-agent-tool-calls-dag-harness 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/parallel-agent-tool-calls-dag-harness-cover.png Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently https://umesh-malik.com/blog/parallel-agent-tool-calls-dag-harness-loop-vs-dag.png Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently https://umesh-malik.com/blog/parallel-agent-tool-calls-dag-harness-budget-pressure.png Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently https://umesh-malik.com/blog/parallel-agent-tool-calls-dag-harness-error-classes.png Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently Diagram contrasting a sequential agent loop of ten model round trips with a two-level dependency graph running nine tool calls concurrently https://umesh-malik.com/blog/reinforcement-fine-tuning-small-models-retrieval 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/reinforcement-fine-tuning-small-models-retrieval-cover.png Reinforcement fine-tuning: a 4B open model matching a frontier LLM on retrieval at a fraction of the cost Reinforcement fine-tuning: a 4B open model matching a frontier LLM on retrieval at a fraction of the cost https://umesh-malik.com/blog/run-cicd-cloudflare-workflows 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/run-cicd-cloudflare-workflows-cover.png Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step https://umesh-malik.com/blog/run-cicd-cloudflare-workflows-yaml-to-ts.png Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step https://umesh-malik.com/blog/run-cicd-cloudflare-workflows-which-meter.png Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step https://umesh-malik.com/blog/run-cicd-cloudflare-workflows-ceilings.png Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step Diagram of a CI pipeline expressed as a Cloudflare Workflow: one cached install step fanning out to four parallel checks and joining into a deploy step https://umesh-malik.com/blog/rust-llm-policy-ai-generated-code 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/rust-llm-policy-ai-generated-code-cover.png Rust LLM policy: AI allowed for review and analysis, banned for creating code and prose Rust LLM policy: AI allowed for review and analysis, banned for creating code and prose https://umesh-malik.com/blog/agent-harness-design-arc-agi-3 2026-08-05 monthly 0.8 https://umesh-malik.com/blog/agent-harness-design-arc-agi-3-cover.png Agent harness design diagram: a model's reasoning discarded each turn versus reasoning retained and context compacted across turns Agent harness design diagram: a model's reasoning discarded each turn versus reasoning retained and context compacted across turns https://umesh-malik.com/blog/insider-threat-offboarding-controls 2026-08-05 monthly 0.8 https://umesh-malik.com/blog/insider-threat-offboarding-controls-cover.png Cover card titled The Exit, subtitled Badge returned. Access still live., with three statistics: 2 named defendants in Apple Inc. v. Liu, 11 additional former employees named in the filing, and reasonable measures as the statutory test for a trade secret Cover card titled The Exit, subtitled Badge returned. Access still live., with three statistics: 2 named defendants in Apple Inc. v. Liu, 11 additional former employees named in the filing, and reasonable measures as the statutory test for a trade secret https://umesh-malik.com/blog/llm-abuse-detection-openai-scam-network 2026-08-05 monthly 0.8 https://umesh-malik.com/blog/llm-abuse-detection-openai-scam-network-cover.png LLM abuse detection diagram contrasting per-message content classification with account-level behavioural correlation across a session corpus LLM abuse detection diagram contrasting per-message content classification with account-level behavioural correlation across a session corpus https://umesh-malik.com/blog/mcp-write-controls-cloudflare-writeguard 2026-08-05 monthly 0.8 https://umesh-malik.com/blog/mcp-write-controls-cloudflare-writeguard-cover.png MCP write controls diagram: a tool call passing through a policy layer that allows, labels, or blocks it before the handler runs MCP write controls diagram: a tool call passing through a policy layer that allows, labels, or blocks it before the handler runs https://umesh-malik.com/blog/secure-mcp-write-tools-writeguard 2026-08-05 monthly 0.8 https://umesh-malik.com/blog/secure-mcp-write-tools-writeguard-cover.png Cover card titled The Gate, subtitled Thousands of tickets. One agent., with three statistics: 4 risk tiers, Cloudflare's MCP portal growing 13 to 27 servers, and tool annotations marked untrusted per the MCP spec Cover card titled The Gate, subtitled Thousands of tickets. One agent., with three statistics: 4 risk tiers, Cloudflare's MCP portal growing 13 to 27 servers, and tool annotations marked untrusted per the MCP spec https://umesh-malik.com/blog/cloudflare-wallets-x402-agent-payments 2026-08-04 monthly 0.8 https://umesh-malik.com/blog/cloudflare-wallets-x402-agent-payments-cover.png Diagram of an AI agent paying an API over the x402 HTTP 402 protocol using a Cloudflare Virtual Wallet Diagram of an AI agent paying an API over the x402 HTTP 402 protocol using a Cloudflare Virtual Wallet https://umesh-malik.com/blog/fake-cve-reports-ai-slop-sqlite 2026-08-04 monthly 0.8 https://umesh-malik.com/blog/fake-cve-reports-ai-slop-sqlite-cover.png A fabricated CVE advisory flowing unverified into the public vulnerability database pipeline A fabricated CVE advisory flowing unverified into the public vulnerability database pipeline https://umesh-malik.com/blog/llm-eval-framework-smevals 2026-08-04 monthly 0.8 https://umesh-malik.com/blog/llm-eval-framework-smevals-cover.png Diagram of an LLM eval framework pipeline: tasks and configs produce runs, graders apply checks to produce grades Diagram of an LLM eval framework pipeline: tasks and configs produce runs, graders apply checks to produce grades https://umesh-malik.com/blog/production-grade-ai-agents-vibe-to-live-gap 2026-08-04 monthly 0.8 https://umesh-malik.com/blog/production-grade-ai-agents-vibe-to-live-gap-cover.png Production-grade AI agents — the funnel from 353,000 course registrations down to 6,000 shipped agent projects Production-grade AI agents — the funnel from 353,000 course registrations down to 6,000 shipped agent projects https://umesh-malik.com/blog/run-70b-llm-on-4gb-gpu-airllm 2026-08-04 monthly 0.8 https://umesh-malik.com/blog/run-70b-llm-on-4gb-gpu-airllm-cover.png Layer-by-layer LLM inference streaming one transformer layer at a time from disk onto a 4GB GPU Layer-by-layer LLM inference streaming one transformer layer at a time from disk onto a 4GB GPU https://umesh-malik.com/blog/agent-to-human-delegation 2026-08-03 monthly 0.8 https://umesh-malik.com/blog/agent-to-human-delegation-cover.png Agent-to-human delegation — an AI agent handing a task to a human coworker in a chat tool Agent-to-human delegation — an AI agent handing a task to a human coworker in a chat tool https://umesh-malik.com/blog/explorative-modeling-train-on-best-of-k 2026-08-02 monthly 0.8 https://umesh-malik.com/blog/explorative-modeling-train-on-best-of-k-cover.png Explorative modeling diagram: K candidate generations compared against real data, with only the closest match receiving gradients Explorative modeling diagram: K candidate generations compared against real data, with only the closest match receiving gradients https://umesh-malik.com/blog/deepseek-v4-flash-0731-benchmarks 2026-08-01 monthly 0.8 https://umesh-malik.com/blog/deepseek-v4-flash-0731-benchmarks-cover.png DeepSeek V4 Flash 0731 agent benchmark results compared against the preview build and V4-Pro Preview DeepSeek V4 Flash 0731 agent benchmark results compared against the preview build and V4-Pro Preview https://umesh-malik.com/blog/increase-seo-traffic-ai-era-techniques 2026-07-29 monthly 0.8 https://umesh-malik.com/blog/increase-seo-traffic-ai-era-techniques-cover.png How to increase SEO traffic in the AI era — a hands-on field guide of genuine, repeatable techniques to earn clicks and AI citations in 2026 How to increase SEO traffic in the AI era — a hands-on field guide of genuine, repeatable techniques to earn clicks and AI citations in 2026 https://umesh-malik.com/blog/seo-in-the-ai-era-geo-playbook 2026-07-27 monthly 0.8 https://umesh-malik.com/blog/seo-in-the-ai-era-geo-playbook-cover.png SEO in the AI era — the shift from ranking blue links to being cited inside AI-generated answers on ChatGPT, Perplexity and Google AI Mode SEO in the AI era — the shift from ranking blue links to being cited inside AI-generated answers on ChatGPT, Perplexity and Google AI Mode https://umesh-malik.com/blog/claude-opus-5-guide 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/claude-opus-5-guide-cover.png Cover for the Claude Opus 5 guide: benchmarks, the $5/$25 cost math, effort levels, and the two API breaking changes in the Opus 4.8 migration Cover for the Claude Opus 5 guide: benchmarks, the $5/$25 cost math, effort levels, and the two API breaking changes in the Opus 4.8 migration https://umesh-malik.com/blog/rag-chatbot-nextjs-guide 2026-07-21 monthly 0.8 https://umesh-malik.com/blog/rag-chatbot-nextjs-guide-cover.png A RAG chatbot in Next.js: embed the query, search pgvector, augment the prompt, stream a cited answer A RAG chatbot in Next.js: embed the query, search pgvector, augment the prompt, stream a cited answer https://umesh-malik.com/blog/react-server-components-guide 2026-07-21 monthly 0.8 https://umesh-malik.com/blog/react-server-components-guide-cover.png React Server Components and the use client boundary: server graph, client graph, and where the line is drawn React Server Components and the use client boundary: server graph, client graph, and where the line is drawn https://umesh-malik.com/blog/vercel-ai-sdk-production-guide 2026-07-21 monthly 0.8 https://umesh-malik.com/blog/vercel-ai-sdk-production-guide-cover.png The production layer of a Vercel AI SDK app: streaming, tool-calling, abort, rate limiting, and cost control The production layer of a Vercel AI SDK app: streaming, tool-calling, abort, rate limiting, and cost control https://umesh-malik.com/blog/kimi-k3-vs-claude-fable-5 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/kimi-k3-vs-fable-5-cover.png Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins https://umesh-malik.com/blog/kimi-k3-benchmark-scorecard.png Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins https://umesh-malik.com/blog/kimi-k3-cost-comparison.png Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins https://umesh-malik.com/blog/kimi-k3-architecture.png Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins Kimi K3 vs Claude Fable 5 head-to-head showing benchmarks, pricing, and where the open 2.8T model wins https://umesh-malik.com/blog/streaming-html-out-of-order-without-javascript 2026-07-14 monthly 0.8 https://umesh-malik.com/blog/streaming-html-out-of-order-without-javascript-cover.png Cover showing an HTML document streaming placeholder markers that get filled out of order by template-for elements, with no JavaScript Cover showing an HTML document streaming placeholder markers that get filled out of order by template-for elements, with no JavaScript https://umesh-malik.com/blog/chatgpt-apps-sdk-super-app-guide 2026-07-11 monthly 0.8 https://umesh-malik.com/blog/chatgpt-apps-sdk-cover.png ChatGPT super app reform showing the Apps SDK built on MCP with inline app UIs and the App Directory ChatGPT super app reform showing the Apps SDK built on MCP with inline app UIs and the App Directory https://umesh-malik.com/blog/chatgpt-super-app-anatomy.png ChatGPT super app reform showing the Apps SDK built on MCP with inline app UIs and the App Directory ChatGPT super app reform showing the Apps SDK built on MCP with inline app UIs and the App Directory https://umesh-malik.com/blog/chatgpt-apps-sdk-architecture.png ChatGPT super app reform showing the Apps SDK built on MCP with inline app UIs and the App Directory ChatGPT super app reform showing the Apps SDK built on MCP with inline app UIs and the App Directory https://umesh-malik.com/blog/gpt-5-6-sol-vs-terra-vs-luna 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/gpt-5-6-sol-terra-luna-cover.png GPT-5.6 Sol vs Terra vs Luna comparison showing price, coding strength, and cost per task GPT-5.6 Sol vs Terra vs Luna comparison showing price, coding strength, and cost per task https://umesh-malik.com/blog/gpt-5-6-routing-map.png GPT-5.6 Sol vs Terra vs Luna comparison showing price, coding strength, and cost per task GPT-5.6 Sol vs Terra vs Luna comparison showing price, coding strength, and cost per task https://umesh-malik.com/blog/openai-gpt-5-6-sol-terra-luna-guide 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/gpt-5-6-cover.png OpenAI GPT-5.6 family showing Sol, Terra, and Luna tiers with benchmarks, pricing, and 1.05M context OpenAI GPT-5.6 family showing Sol, Terra, and Luna tiers with benchmarks, pricing, and 1.05M context https://umesh-malik.com/blog/gpt-5-6-family-tiers.png OpenAI GPT-5.6 family showing Sol, Terra, and Luna tiers with benchmarks, pricing, and 1.05M context OpenAI GPT-5.6 family showing Sol, Terra, and Luna tiers with benchmarks, pricing, and 1.05M context https://umesh-malik.com/blog/gpt-5-6-coding-benchmarks.png OpenAI GPT-5.6 family showing Sol, Terra, and Luna tiers with benchmarks, pricing, and 1.05M context OpenAI GPT-5.6 family showing Sol, Terra, and Luna tiers with benchmarks, pricing, and 1.05M context https://umesh-malik.com/blog/build-enterprise-ai-agents-free 2026-07-08 monthly 0.8 https://umesh-malik.com/blog/build-enterprise-ai-agents-free-cover.png Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted https://umesh-malik.com/blog/free-enterprise-agents-stack.card.png Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted https://umesh-malik.com/blog/free-enterprise-agents-rag-precision.card.png Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted https://umesh-malik.com/blog/free-enterprise-agents-cost.card.png Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted https://umesh-malik.com/blog/http-query-method-rfc-10008-guide 2026-07-03 monthly 0.8 https://umesh-malik.com/blog/http-query-method-rfc-10008-guide-cover.png Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers https://umesh-malik.com/blog/http-query-method-gap.png Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers https://umesh-malik.com/blog/http-query-standards-timeline.png Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers https://umesh-malik.com/blog/http-query-caching-flow.png Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers https://umesh-malik.com/blog/http-query-adoption-map.png Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers Cover for the HTTP QUERY method guide: RFC 10008's safe, idempotent, cacheable request method with a body, explained for REST API designers https://umesh-malik.com/blog/claude-fable-5-guide 2026-07-02 monthly 0.8 https://umesh-malik.com/blog/claude-fable-5-guide-cover.png Cover for the Claude Fable 5 deep-dive guide: capabilities, the $10/$50 cost math, API behavior, and when to use Anthropic's most capable model Cover for the Claude Fable 5 deep-dive guide: capabilities, the $10/$50 cost math, API behavior, and when to use Anthropic's most capable model https://umesh-malik.com/blog/claude-sonnet-5-guide 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/claude-sonnet-5-guide-cover.png Cover for the Claude Sonnet 5 guide: benchmarks, pros and cons, use cases, and cost math versus Sonnet 4.6 and Opus 4.8 Cover for the Claude Sonnet 5 guide: benchmarks, pros and cons, use cases, and cost math versus Sonnet 4.6 and Opus 4.8 https://umesh-malik.com/blog/claude-code-auto-mode-production-field-report 2026-06-25 monthly 0.8 https://umesh-malik.com/blog/claude-code-auto-mode-production-field-report-cover.png Field report cover: a week of running Claude Code auto mode in production, showing what it nailed, where it broke, and the cost Field report cover: a week of running Claude Code auto mode in production, showing what it nailed, where it broke, and the cost https://umesh-malik.com/blog/claude-code-vs-cursor-production-work-2026 2026-06-25 monthly 0.8 https://umesh-malik.com/blog/claude-code-vs-cursor-production-work-2026-cover.png Decision matrix cover comparing Claude Code, Cursor, and GitHub Copilot for real production work in 2026 Decision matrix cover comparing Claude Code, Cursor, and GitHub Copilot for real production work in 2026 https://umesh-malik.com/blog/use-claude-code-codex-for-free 2026-06-23 monthly 0.8 https://umesh-malik.com/blog/use-claude-code-codex-for-free-cover.png Cover comparing free and paid access to Claude Code, Codex, Gemini CLI, and GitHub Copilot AI coding agents in 2026 Cover comparing free and paid access to Claude Code, Codex, Gemini CLI, and GitHub Copilot AI coding agents in 2026 https://umesh-malik.com/blog/fastapi-spa-app-frontend-explained 2026-06-21 monthly 0.8 https://umesh-malik.com/blog/fastapi-spa-app-frontend-explained-cover.png Cover showing FastAPI's new app.frontend() method mounting a built SPA directory alongside API routes Cover showing FastAPI's new app.frontend() method mounting a built SPA directory alongside API routes https://umesh-malik.com/blog/agentic-browsing-pagespeed-ai-ready 2026-06-19 monthly 0.8 https://umesh-malik.com/blog/agentic-browsing-pagespeed-ai-ready-cover.png The Agentic Browsing category in PageSpeed Insights scoring a website for AI agents alongside Performance, Accessibility, Best Practices and SEO The Agentic Browsing category in PageSpeed Insights scoring a website for AI agents alongside Performance, Accessibility, Best Practices and SEO https://umesh-malik.com/blog/agentic-browsing-pagespeed-ai-ready-lighthouse-result.png The Agentic Browsing category in PageSpeed Insights scoring a website for AI agents alongside Performance, Accessibility, Best Practices and SEO The Agentic Browsing category in PageSpeed Insights scoring a website for AI agents alongside Performance, Accessibility, Best Practices and SEO https://umesh-malik.com/blog/autonomous-ai-agents-production-gap-2026 2026-06-14 monthly 0.8 https://umesh-malik.com/blog/autonomous-ai-agents-production-gap-2026-cover.png Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 https://umesh-malik.com/blog/autonomous-agents-production-gap.card.png Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 https://umesh-malik.com/blog/autonomous-agents-compounding-error.card.png Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 https://umesh-malik.com/blog/autonomous-agents-autonomy-spectrum.card.png Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 https://umesh-malik.com/blog/autonomous-agents-lassie-case-study.card.png Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 Editorial cover: the demo-to-production gap for autonomous AI agents in 2026 https://umesh-malik.com/blog/claude-fable-5-streaming-microservice-one-day 2026-06-13 monthly 0.8 https://umesh-malik.com/blog/claude-fable-5-streaming-microservice-one-day-cover.png Cover showing the one-day flow of Claude Fable 5 building a streaming microservice: spec, infrastructure, code, and ship Cover showing the one-day flow of Claude Fable 5 building a streaming microservice: spec, infrastructure, code, and ship https://umesh-malik.com/blog/fable-auto-mode-timeline.png Cover showing the one-day flow of Claude Fable 5 building a streaming microservice: spec, infrastructure, code, and ship Cover showing the one-day flow of Claude Fable 5 building a streaming microservice: spec, infrastructure, code, and ship https://umesh-malik.com/blog/fable-streaming-architecture.png Cover showing the one-day flow of Claude Fable 5 building a streaming microservice: spec, infrastructure, code, and ship Cover showing the one-day flow of Claude Fable 5 building a streaming microservice: spec, infrastructure, code, and ship https://umesh-malik.com/blog/build-rag-pipeline-from-scratch 2026-06-08 monthly 0.8 https://umesh-malik.com/blog/build-rag-pipeline-from-scratch-cover.png The stages of a production retrieval-augmented generation pipeline The stages of a production retrieval-augmented generation pipeline https://umesh-malik.com/blog/cursor-vs-claude-code-vs-copilot 2026-06-08 monthly 0.8 https://umesh-malik.com/blog/cursor-vs-claude-code-vs-copilot-cover.png Comparison of Cursor, Claude Code, and GitHub Copilot AI coding tools Comparison of Cursor, Claude Code, and GitHub Copilot AI coding tools https://umesh-malik.com/blog/deploy-mcp-server-cloudflare-workers 2026-06-08 monthly 0.8 https://umesh-malik.com/blog/deploy-mcp-server-cloudflare-workers-cover.png Deploying a Model Context Protocol server to the edge on Cloudflare Workers Deploying a Model Context Protocol server to the edge on Cloudflare Workers https://umesh-malik.com/blog/how-to-build-mcp-server 2026-07-21 monthly 0.8 https://umesh-malik.com/blog/how-to-build-mcp-server-cover.png Architecture of a production Model Context Protocol server on Cloudflare Workers Architecture of a production Model Context Protocol server on Cloudflare Workers https://umesh-malik.com/blog/how-to-write-claude-md 2026-06-08 monthly 0.8 https://umesh-malik.com/blog/how-to-write-claude-md-cover.png Anatomy of an effective CLAUDE.md project context file for Claude Code Anatomy of an effective CLAUDE.md project context file for Claude Code https://umesh-malik.com/blog/claude-swap-multi-account-switcher-guide 2026-07-27 monthly 0.8 https://umesh-malik.com/blog/claude-swap-multi-account-switcher-guide-cover.png claude-swap cover showing the four-step flow of switching Claude Code accounts: add account, switch, restore credentials, restart claude-swap cover showing the four-step flow of switching Claude Code accounts: add account, switch, restore credentials, restart https://umesh-malik.com/blog/claude-swap-flow.png claude-swap cover showing the four-step flow of switching Claude Code accounts: add account, switch, restore credentials, restart claude-swap cover showing the four-step flow of switching Claude Code accounts: add account, switch, restore credentials, restart https://umesh-malik.com/blog/claude-swap-storage.png claude-swap cover showing the four-step flow of switching Claude Code accounts: add account, switch, restore credentials, restart claude-swap cover showing the four-step flow of switching Claude Code accounts: add account, switch, restore credentials, restart https://umesh-malik.com/blog/claude-code-leak-march-2026 2026-04-02 monthly 0.8 https://umesh-malik.com/blog/claude-code-leak-cover.png Claude Code source map leak visualization showing 512K lines of TypeScript escaping a code window with key stats Claude Code source map leak visualization showing 512K lines of TypeScript escaping a code window with key stats https://umesh-malik.com/blog/claude-code-leak-pipeline.png Claude Code source map leak visualization showing 512K lines of TypeScript escaping a code window with key stats Claude Code source map leak visualization showing 512K lines of TypeScript escaping a code window with key stats https://umesh-malik.com/blog/axios-compromised-npm-cross-platform-rat 2026-03-31 monthly 0.8 https://umesh-malik.com/blog/axios-compromised-npm-cover.png Editorial cover: the Axios npm supply-chain compromise (1.14.1 and 0.30.4) Editorial cover: the Axios npm supply-chain compromise (1.14.1 and 0.30.4) https://umesh-malik.com/blog/nvidia-openclaw-strategy-ai-agent-plan 2026-03-17 monthly 0.8 https://umesh-malik.com/blog/nvidia-openclaw-cover.png Nvidia OpenClaw strategy cover showing task assignment, agent execution, guardrails, and enterprise runtime control Nvidia OpenClaw strategy cover showing task assignment, agent execution, guardrails, and enterprise runtime control https://umesh-malik.com/blog/nvidia-openclaw-shift.png Nvidia OpenClaw strategy cover showing task assignment, agent execution, guardrails, and enterprise runtime control Nvidia OpenClaw strategy cover showing task assignment, agent execution, guardrails, and enterprise runtime control https://umesh-malik.com/blog/nvidia-agent-runtime-stack.png Nvidia OpenClaw strategy cover showing task assignment, agent execution, guardrails, and enterprise runtime control Nvidia OpenClaw strategy cover showing task assignment, agent execution, guardrails, and enterprise runtime control https://umesh-malik.com/blog/chatgpt-adult-mode-delay-guide 2026-03-16 monthly 0.8 https://umesh-malik.com/blog/chatgpt-adult-mode-cover.png ChatGPT adult mode cover showing age prediction, adult verification, text-only scope, and safety guardrails ChatGPT adult mode cover showing age prediction, adult verification, text-only scope, and safety guardrails https://umesh-malik.com/blog/chatgpt-adult-mode-scope.png ChatGPT adult mode cover showing age prediction, adult verification, text-only scope, and safety guardrails ChatGPT adult mode cover showing age prediction, adult verification, text-only scope, and safety guardrails https://umesh-malik.com/blog/chatgpt-adult-mode-tension-map.png ChatGPT adult mode cover showing age prediction, adult verification, text-only scope, and safety guardrails ChatGPT adult mode cover showing age prediction, adult verification, text-only scope, and safety guardrails https://umesh-malik.com/blog/chatgpt-interactive-math-science-visuals-guide 2026-03-12 monthly 0.8 https://umesh-malik.com/blog/chatgpt-learning-visuals-cover.png ChatGPT interactive learning cover showing math and science concepts becoming visual and interactive inside ChatGPT ChatGPT interactive learning cover showing math and science concepts becoming visual and interactive inside ChatGPT https://umesh-malik.com/blog/chatgpt-learning-loop.png ChatGPT interactive learning cover showing math and science concepts becoming visual and interactive inside ChatGPT ChatGPT interactive learning cover showing math and science concepts becoming visual and interactive inside ChatGPT https://umesh-malik.com/blog/chatgpt-math-science-topic-map.png ChatGPT interactive learning cover showing math and science concepts becoming visual and interactive inside ChatGPT ChatGPT interactive learning cover showing math and science concepts becoming visual and interactive inside ChatGPT https://umesh-malik.com/blog/anthropic-code-review-claude-code-guide 2026-03-10 monthly 0.8 https://umesh-malik.com/blog/anthropic-code-review-cover.png Anthropic Code Review cover showing multi-agent pull request review across repository context and custom review rules Anthropic Code Review cover showing multi-agent pull request review across repository context and custom review rules https://umesh-malik.com/blog/anthropic-code-review-loop.png Anthropic Code Review cover showing multi-agent pull request review across repository context and custom review rules Anthropic Code Review cover showing multi-agent pull request review across repository context and custom review rules https://umesh-malik.com/blog/anthropic-code-review-stack.png Anthropic Code Review cover showing multi-agent pull request review across repository context and custom review rules Anthropic Code Review cover showing multi-agent pull request review across repository context and custom review rules https://umesh-malik.com/blog/agentic-ai-enterprise-security-model 2026-03-09 monthly 0.8 https://umesh-malik.com/blog/agentic-ai-enterprise-security-cover.png Agentic AI enterprise security cover showing identity, prompt injection, policy gates, and observability Agentic AI enterprise security cover showing identity, prompt injection, policy gates, and observability https://umesh-malik.com/blog/agentic-ai-security-model-shift.png Agentic AI enterprise security cover showing identity, prompt injection, policy gates, and observability Agentic AI enterprise security cover showing identity, prompt injection, policy gates, and observability https://umesh-malik.com/blog/agentic-ai-control-plane.png Agentic AI enterprise security cover showing identity, prompt injection, policy gates, and observability Agentic AI enterprise security cover showing identity, prompt injection, policy gates, and observability https://umesh-malik.com/blog/openai-gpt-5-4-complete-guide 2026-08-06 monthly 0.8 https://umesh-malik.com/blog/gpt-5-4-cover.png OpenAI GPT-5.4 overview showing professional work, coding, computer use, and 1M context OpenAI GPT-5.4 overview showing professional work, coding, computer use, and 1M context https://umesh-malik.com/blog/gpt-5-4-capability-stack.png OpenAI GPT-5.4 overview showing professional work, coding, computer use, and 1M context OpenAI GPT-5.4 overview showing professional work, coding, computer use, and 1M context https://umesh-malik.com/blog/gpt-5-4-model-selection-map.png OpenAI GPT-5.4 overview showing professional work, coding, computer use, and 1M context OpenAI GPT-5.4 overview showing professional work, coding, computer use, and 1M context https://umesh-malik.com/blog/openai-gpt-5-3-instant-fewer-refusals-better-answers 2026-03-04 monthly 0.8 https://umesh-malik.com/blog/gpt-5-3-instant-cover.png OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone https://umesh-malik.com/blog/gpt-5-3-instant-improvement-map.png OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone https://umesh-malik.com/blog/gpt-5-3-instant-hallucination-reduction.png OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone https://umesh-malik.com/blog/gpt-5-3-instant-vs-5-2-comparison.png OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone OpenAI GPT-5.3 Instant overview showing three key improvements: fewer refusals, better web answers, and smoother conversational tone https://umesh-malik.com/blog/deepseek-v4-release-challenge-us-ai-rivals 2026-03-01 monthly 0.8 https://umesh-malik.com/blog/deepseek-v4-release-cover.png DeepSeek V4 launch preview showing AI model race between China-first chips and U.S. rivals DeepSeek V4 launch preview showing AI model race between China-first chips and U.S. rivals https://umesh-malik.com/blog/deepseek-v4-competition-map.png DeepSeek V4 launch preview showing AI model race between China-first chips and U.S. rivals DeepSeek V4 launch preview showing AI model race between China-first chips and U.S. rivals https://umesh-malik.com/blog/rag-vs-fine-tuning-llms-2026 2026-02-28 monthly 0.8 https://umesh-malik.com/blog/rag-vs-fine-tuning-llms-2026-cover.png RAG vs fine-tuning architecture comparison for LLMs RAG vs fine-tuning architecture comparison for LLMs https://umesh-malik.com/blog/rag-vs-fine-tuning-architecture.png RAG vs fine-tuning architecture comparison for LLMs RAG vs fine-tuning architecture comparison for LLMs https://umesh-malik.com/blog/rag-vs-fine-tuning-decision-tree.png RAG vs fine-tuning architecture comparison for LLMs RAG vs fine-tuning architecture comparison for LLMs https://umesh-malik.com/blog/figma-codex-react-2026 2026-02-27 monthly 0.8 https://umesh-malik.com/blog/figma-codex-react-cover.png Editorial cover: converting Figma designs to React with OpenAI Codex Editorial cover: converting Figma designs to React with OpenAI Codex https://umesh-malik.com/blog/cloudflare-vinext-next-js-vite-revolution 2026-02-25 monthly 0.8 https://umesh-malik.com/blog/cloudflare-vinext-cover.png A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster https://umesh-malik.com/blog/vinext-ai-development-comparison.png A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster https://umesh-malik.com/blog/vinext-traffic-aware-prerendering.png A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster https://umesh-malik.com/blog/vinext-dx-performance-evolution.png A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster A lightning bolt splitting between Next.js and Vite logos, symbolizing Cloudflare's revolutionary Vinext framework that's 4.4x faster https://umesh-malik.com/blog/anthropic-detecting-preventing-distillation-attacks 2026-02-24 monthly 0.8 https://umesh-malik.com/blog/distillation-attacks-cover.png A glowing AI brain being extracted through a network of fraudulent connections representing the massive distillation attack on Claude A glowing AI brain being extracted through a network of fraudulent connections representing the massive distillation attack on Claude https://umesh-malik.com/blog/hydra-network-proxy-attacks.png A glowing AI brain being extracted through a network of fraudulent connections representing the massive distillation attack on Claude A glowing AI brain being extracted through a network of fraudulent connections representing the massive distillation attack on Claude https://umesh-malik.com/blog/ai-national-security-threat.png A glowing AI brain being extracted through a network of fraudulent connections representing the massive distillation attack on Claude A glowing AI brain being extracted through a network of fraudulent connections representing the massive distillation attack on Claude https://umesh-malik.com/blog/local-llm-coding-revolution-qwen3-coder-desktop 2026-02-22 monthly 0.8 https://umesh-malik.com/blog/local-llm-coding-cover.png A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding https://umesh-malik.com/blog/local-llm-moe-architecture.png A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding https://umesh-malik.com/blog/local-llm-cost-breakeven.png A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding https://umesh-malik.com/blog/local-llm-quantization-tradeoff.png A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding https://umesh-malik.com/blog/local-llm-hardware-tiers.png A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding A desktop workstation running a local LLM for coding — 80 billion parameters, 3 billion active — representing the shift from cloud AI to local AI coding https://umesh-malik.com/blog/spec-driven-development-ai-agents-addy-osmani 2026-02-21 monthly 0.8 https://umesh-malik.com/blog/spec-driven-dev-cover.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/spec-driven-dev-the-gap.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/spec-driven-dev-curse-of-instructions.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/spec-driven-dev-three-tier-boundaries.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/spec-driven-dev-four-phase-workflow.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/spec-driven-dev-vibe-vs-production.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/spec-driven-dev-engineering-manager.png Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document Three AI coding agents — database, security, and test — sitting at terminal screens writing and reviewing code, all guided by a central SPEC.md specification document https://umesh-malik.com/blog/nodejs-memory-cut-in-half-pointer-compression 2026-02-18 monthly 0.8 https://umesh-malik.com/blog/nodejs-memory-cut-in-half-cover.png Editorial cover: Node.js V8 pointer compression cuts heap memory in half Editorial cover: Node.js V8 pointer compression cuts heap memory in half https://umesh-malik.com/blog/agents-md-ai-coding-agents-study 2026-02-17 monthly 0.8 https://umesh-malik.com/blog/agents-md-cover.png AGENTS.md study results showing a document icon next to a declining performance chart from ETH Zurich's 138-repo benchmark AGENTS.md study results showing a document icon next to a declining performance chart from ETH Zurich's 138-repo benchmark https://umesh-malik.com/blog/agents-md-format-war.png AGENTS.md study results showing a document icon next to a declining performance chart from ETH Zurich's 138-repo benchmark AGENTS.md study results showing a document icon next to a declining performance chart from ETH Zurich's 138-repo benchmark https://umesh-malik.com/blog/docker-swarm-vs-kubernetes-166-dollar-reality-check 2026-02-17 monthly 0.8 https://umesh-malik.com/blog/docker-swarm-vs-k8s-cover.png Docker Swarm versus Kubernetes cost comparison: $166 per year simple setup versus $200K complex infrastructure Docker Swarm versus Kubernetes cost comparison: $166 per year simple setup versus $200K complex infrastructure https://umesh-malik.com/blog/docker-swarm-yaml-comparison.png Docker Swarm versus Kubernetes cost comparison: $166 per year simple setup versus $200K complex infrastructure Docker Swarm versus Kubernetes cost comparison: $166 per year simple setup versus $200K complex infrastructure https://umesh-malik.com/blog/seedance-2-hollywood-ai-copyright-crisis 2026-02-16 monthly 0.8 https://umesh-malik.com/blog/seedance-cover.png The Seedance 2.0 crisis: Hollywood confrontation with AI-generated video showing film reel colliding with AI The Seedance 2.0 crisis: Hollywood confrontation with AI-generated video showing film reel colliding with AI https://umesh-malik.com/blog/seedance-competitive-landscape.png The Seedance 2.0 crisis: Hollywood confrontation with AI-generated video showing film reel colliding with AI The Seedance 2.0 crisis: Hollywood confrontation with AI-generated video showing film reel colliding with AI https://umesh-malik.com/blog/ai-agent-attacks-developer-matplotlib-open-source 2026-02-15 monthly 0.8 https://umesh-malik.com/blog/ai-agent-matplotlib-cover.png Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer https://umesh-malik.com/blog/ai-agent-retaliation.webp Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer https://umesh-malik.com/blog/ai-code-hacker-dark.webp Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer https://umesh-malik.com/blog/ai-future-dark.webp Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer Editorial cover: an AI agent's autonomous retaliation against an open-source maintainer https://umesh-malik.com/blog/nodejs-backend-for-frontend-developers 2026-02-10 monthly 0.8 https://umesh-malik.com/blog/nodejs-backend-cover.png Node.js backend architecture showing Express server, middleware chain, REST API endpoints, and database layer Node.js backend architecture showing Express server, middleware chain, REST API endpoints, and database layer https://umesh-malik.com/blog/frontend-career-growth-junior-to-senior 2026-01-08 monthly 0.8 https://umesh-malik.com/blog/frontend-career-cover.png Frontend career growth timeline showing progression from Associate Engineer through to SDE-2 at Expedia Group Frontend career growth timeline showing progression from Associate Engineer through to SDE-2 at Expedia Group https://umesh-malik.com/blog/core-web-vitals-optimization-guide 2025-11-12 monthly 0.8 https://umesh-malik.com/blog/core-web-vitals-cover.png Core Web Vitals dashboard showing three gauge meters for LCP, INP, and CLS with before and after optimization results Core Web Vitals dashboard showing three gauge meters for LCP, INP, and CLS with before and after optimization results https://umesh-malik.com/blog/developer-productivity-tools-senior-engineers 2025-09-05 monthly 0.8 https://umesh-malik.com/blog/dev-productivity-cover.png Developer productivity workspace showing a terminal, editor, and organized workflow system for senior engineers Developer productivity workspace showing a terminal, editor, and organized workflow system for senior engineers https://umesh-malik.com/blog/frontend-testing-strategies-2025 2025-07-10 monthly 0.8 https://umesh-malik.com/blog/frontend-testing-cover.png Frontend testing trophy model with layers for static analysis, integration tests, unit tests, and E2E tests Frontend testing trophy model with layers for static analysis, integration tests, unit tests, and E2E tests https://umesh-malik.com/blog/tailwindcss-v4-migration-guide 2026-07-21 monthly 0.8 https://umesh-malik.com/blog/tailwindcss-v4-cover.png TailwindCSS v4 migration showing the shift from JavaScript config to CSS-first @theme configuration TailwindCSS v4 migration showing the shift from JavaScript config to CSS-first @theme configuration https://umesh-malik.com/blog/javascript-es2024-features-you-should-know 2025-03-15 monthly 0.8 https://umesh-malik.com/blog/javascript-es2024-cover.png JavaScript ES2024 features overview showing Object.groupBy, Promise.withResolvers, Unicode strings, and RegExp v flag JavaScript ES2024 features overview showing Object.groupBy, Promise.withResolvers, Unicode strings, and RegExp v flag https://umesh-malik.com/blog/typescript-utility-types-complete-guide 2026-07-20 monthly 0.8 https://umesh-malik.com/blog/typescript-utility-cover.png TypeScript utility types overview showing Partial, Required, Pick, Omit, and Record with code examples TypeScript utility types overview showing Partial, Required, Pick, Omit, and Record with code examples https://umesh-malik.com/blog/react-performance-optimization-techniques 2024-11-20 monthly 0.8 https://umesh-malik.com/blog/react-performance-cover.png React performance optimization showing 10 techniques including memoization, code splitting, and virtualization React performance optimization showing 10 techniques including memoization, code splitting, and virtualization https://umesh-malik.com/blog/sveltekit-vs-nextjs-comparison 2024-10-10 monthly 0.8 https://umesh-malik.com/blog/sveltekit-vs-nextjs-cover.png SvelteKit versus Next.js framework comparison across performance, developer experience, routing, and ecosystem SvelteKit versus Next.js framework comparison across performance, developer experience, routing, and ecosystem