--- author: "Umesh Malik" canonical: "https://umesh-malik.com/blog/category/ai-engineering" description: "Explore AI Engineering articles and tutorials by Umesh Malik — AI Engineer, LLM & GenAI Developer. Learn AI Engineering best practices, practical tips, and in-depth guides." title: "Umesh Malik's Blog - AI Engineering Articles | AI Engineering Tutorials" tokens: 2950 generator: "scripts/generate-page-markdown.mjs" --- [← Back to Blog](https://umesh-malik.com/blog) # AI Engineering 22 articles [![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-cover.png) AI Engineering • Aug 25, 2026 ### Fix the PCIe Bottleneck in AI Training: How Built-in NICs Work Fix the PCIe bottleneck in AI training with built-in NICs. Meta's MTIA 300 reclaims 1.2 TB/s by eliminating host CPU mediation. 9 min read Read more →](https://umesh-malik.com/blog/eliminate-pcie-bottleneck-ai-training) [![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-cover.png) AI Engineering • Aug 19, 2026 ### Rust GPU Offload Without unsafe: The 400x Mistake That Isn't Safety Rust GPU offload now works without unsafe code. Real benchmarks: 11% faster to 46% slower than CUDA on an H100, and a transfer bug that costs 400x more. 8 min read Read more →](https://umesh-malik.com/blog/rust-safe-gpu-offload-benchmarks) [![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-cover.png) AI Engineering • Aug 16, 2026 ### How DNA wires a brain: 300M bits for 100 trillion connections How DNA wires a brain: a ~10B-bit genome must wire 100 trillion connections. The scheme that closes the gap in 300M bits, and why two simpler plans fail first. 9 min read Read more →](https://umesh-malik.com/blog/how-dna-wires-a-brain) [![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-cover.png) AI Engineering • Aug 12, 2026 ### Agent context compaction: keep what the 150K cutoff drops Agent context compaction drops every block before the summary at 150K tokens. What survives, what instructions silently replaces, and the usage field that lies. 9 min read Read more →](https://umesh-malik.com/blog/agent-context-compaction-what-survives) [![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-cover.png) AI Engineering • Aug 7, 2026 ### Set up AI Gateway for Workers AI: one argument, every call logged AI Gateway for Workers AI is now one argument on env.AI.run. What it logs instantly, why caching stays off until you ask, and the 60-second TTL floor. 10 min read Read more →](https://umesh-malik.com/blog/ai-gateway-for-workers-ai) [![Ads for AI agents: TIME serves markdown with sponsored FAQ blocks to AI crawlers and HTML to humans](https://umesh-malik.com/blog/ads-for-ai-agents-time-markdown-crawlers-cover.png) AI Engineering • Aug 6, 2026 ### Ads for AI Agents: TIME Serves Crawlers a Different Site Ads for AI agents are live. TIME serves crawlers a 42 KB markdown site with sponsored blocks; humans and Googlebot get 1.2 MB of HTML. I measured all of it. 10 min read Read more →](https://umesh-malik.com/blog/ads-for-ai-agents-time-markdown-crawlers) [![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-cover.png) AI Engineering • Aug 6, 2026 ### Run agent tool calls in parallel: 10 turns become 2 DAG levels Run agent tool calls in parallel by swapping the loop for a DAG planner: ten round trips become two levels, plus the cap, budget and critic on top. 10 min read Read more →](https://umesh-malik.com/blog/parallel-agent-tool-calls-dag-harness) [![Agent harness design diagram: a model's reasoning discarded each turn versus reasoning retained and context compacted across turns](https://umesh-malik.com/blog/agent-harness-design-arc-agi-3-cover.png) AI Engineering • Aug 5, 2026 ### Agent Harness Design: Why an ARC-AGI-3 Score Tripled Agent harness design decided a benchmark: OpenAI's ARC-AGI-3 score went 13.3% → 38.3% with zero model changes. What that means for your agent loop. 11 min read Read more →](https://umesh-malik.com/blog/agent-harness-design-arc-agi-3) [![Diagram of an AI agent paying an API over the x402 HTTP 402 protocol using a Cloudflare Virtual Wallet](https://umesh-malik.com/blog/cloudflare-wallets-x402-agent-payments-cover.png) AI Engineering • Aug 4, 2026 ### Cloudflare Wallets and x402: How AI Agents Pay for APIs Cloudflare Wallets and x402 explained: how AI agents get a spending identity, how HTTP 402 payments work, and what breaks when your agent holds a budget. 10 min read Read more →](https://umesh-malik.com/blog/cloudflare-wallets-x402-agent-payments) [![Production-grade AI agents — the funnel from 353,000 course registrations down to 6,000 shipped agent projects](https://umesh-malik.com/blog/production-grade-ai-agents-vibe-to-live-gap-cover.png) AI Engineering • Aug 4, 2026 ### Production-Grade AI Agents: The Vibe-to-Live Gap Production-grade AI agents need more than a working demo. 353,000 signed up to learn agents; 6,000 shipped a capstone. Here's exactly what closes the gap. 11 min read Read more →](https://umesh-malik.com/blog/production-grade-ai-agents-vibe-to-live-gap) [![Agent-to-human delegation — an AI agent handing a task to a human coworker in a chat tool](https://umesh-malik.com/blog/agent-to-human-delegation-cover.png) AI Engineering • Aug 3, 2026 ### Agent-to-Human Delegation: Why Your AI Shouldn't DM Coworkers Agent-to-human delegation is the pattern nobody designed for: your AI messaging a coworker. Why it lands badly, and the handoff rules that fix it. 10 min read Read more →](https://umesh-malik.com/blog/agent-to-human-delegation) [![Explorative modeling diagram: K candidate generations compared against real data, with only the closest match receiving gradients](https://umesh-malik.com/blog/explorative-modeling-train-on-best-of-k-cover.png) AI Engineering • Aug 2, 2026 ### Explorative Modeling: Train on the Best of K Guesses Explorative modeling factors the training loop instead of generation: sample K guesses, train only on the best. 4.1x FLOP efficiency, 1.43 FID on ImageNet. 8 min read Read more →](https://umesh-malik.com/blog/explorative-modeling-train-on-best-of-k) [![A RAG chatbot in Next.js: embed the query, search pgvector, augment the prompt, stream a cited answer](https://umesh-malik.com/blog/rag-chatbot-nextjs-guide-cover.png) AI Engineering • Jul 21, 2026 ### Build a RAG Chatbot in Next.js: Retrieval, Streaming & Citations (2026) Build a RAG chatbot in Next.js with the AI SDK: embed the query, search pgvector, stream a grounded answer with citations, and stop hallucinations. 8 min read Read more →](https://umesh-malik.com/blog/rag-chatbot-nextjs-guide) [![The production layer of a Vercel AI SDK app: streaming, tool-calling, abort, rate limiting, and cost control](https://umesh-malik.com/blog/vercel-ai-sdk-production-guide-cover.png) AI Engineering • Jul 21, 2026 ### Vercel AI SDK in Production: Streaming, Tool-Calling & the Gotchas Nobody Tells You (2026) Vercel AI SDK in production: streaming, tool-calling, aborting generations, error retry UX, rate limiting, and cost control — the layer every tutorial skips. 9 min read Read more →](https://umesh-malik.com/blog/vercel-ai-sdk-production-guide) [![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-cover.png) AI Engineering • Jul 11, 2026 ### ChatGPT Apps SDK and the Super App Reform: How Apps in ChatGPT Work (2026) The ChatGPT Apps SDK explained: how apps in ChatGPT work, why it's built on MCP, who the launch partners are, and how developers build and submit apps. 7 min read Read more →](https://umesh-malik.com/blog/chatgpt-apps-sdk-super-app-guide) [![Editorial cover: build enterprise-grade AI agents for free with open-source MaxKB, $0 API cost, self-hosted](https://umesh-malik.com/blog/build-enterprise-ai-agents-free-cover.png) AI Engineering • Jul 8, 2026 ### How to Build Enterprise-Grade AI Agents for Free (MaxKB, 2026) How to build enterprise-grade AI agents for free in 2026: a hands-on MaxKB + local LLM guide to RAG precision, security, and $0 API cost. 10 min read Read more →](https://umesh-malik.com/blog/build-enterprise-ai-agents-free) [![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-cover.png) AI Engineering • Jun 19, 2026 ### Agentic Browsing in PageSpeed Insights: How to Make Your Website AI-Ready (2026) Agentic browsing in PageSpeed Insights grades your site for AI agents, not just humans. What the category checks, why Google added it, how to pass it. 9 min read Read more →](https://umesh-malik.com/blog/agentic-browsing-pagespeed-ai-ready) [![Editorial cover: the demo-to-production gap for autonomous AI agents in 2026](https://umesh-malik.com/blog/autonomous-ai-agents-production-gap-2026-cover.png) AI Engineering • Jun 14, 2026 ### Why 77% of Autonomous AI Agents Never Reach Production (2026) Only 23% of autonomous AI agents reach production in 2026. The demo-to-production gap, why agents fail, and the playbook the winners actually use. 9 min read Read more →](https://umesh-malik.com/blog/autonomous-ai-agents-production-gap-2026) [![The stages of a production retrieval-augmented generation pipeline](https://umesh-malik.com/blog/build-rag-pipeline-from-scratch-cover.png) AI Engineering • Jun 8, 2026 ### Build a RAG Pipeline From Scratch: Production Patterns That Matter Build a RAG pipeline from scratch: chunking, embeddings, retrieval, reranking, grounded generation, and the production patterns that decide whether it works. 7 min read Read more →](https://umesh-malik.com/blog/build-rag-pipeline-from-scratch) [![Deploying a Model Context Protocol server to the edge on Cloudflare Workers](https://umesh-malik.com/blog/deploy-mcp-server-cloudflare-workers-cover.png) AI Engineering • Jun 8, 2026 ### Deploy an MCP Server on Cloudflare Workers (Free, at the Edge) Deploy an MCP server on Cloudflare Workers: wrangler.toml, the run\_worker\_first model, routing /mcp, local testing, and going live on the free tier. 7 min read Read more →](https://umesh-malik.com/blog/deploy-mcp-server-cloudflare-workers) [![Architecture of a production Model Context Protocol server on Cloudflare Workers](https://umesh-malik.com/blog/how-to-build-mcp-server-cover.png) AI Engineering • Jun 8, 2026 ### How to Build an MCP Server: A Step-by-Step Guide (2026) How to build an MCP server, step by step: JSON-RPC 2.0, the Streamable HTTP transport, typed tools, and agent discovery — from a real one I shipped. 9 min read Read more →](https://umesh-malik.com/blog/how-to-build-mcp-server) [![RAG vs fine-tuning architecture comparison for LLMs](https://umesh-malik.com/blog/rag-vs-fine-tuning-llms-2026-cover.png) AI Engineering • Feb 28, 2026 ### RAG vs Fine-Tuning for LLMs in 2026: A Production Decision Framework With Real Tradeoffs RAG vs fine-tuning for LLMs in 2026: a practical decision framework covering architecture tradeoffs, cost, latency, and when to use each in production. 6 min read Read more →](https://umesh-malik.com/blog/rag-vs-fine-tuning-llms-2026)