AI Engineer Path https://aiengineerpath.dev/ Practical, hands-on guides for backend engineers learning to build with AI. RAG, evals, MCP, production patterns. en-us AI Engineer Path hello@aiengineerpath.dev (AI Engineer Path) Mon, 26 May 2026 09:00:00 +0000 Mon, 26 May 2026 09:00:00 +0000 What LLMs Are: A Backend Engineer's Mental Model https://aiengineerpath.dev/ai-engineering/articles/01-what-llms-are.html https://aiengineerpath.dev/ai-engineering/articles/01-what-llms-are.html Mon, 26 May 2026 09:00:00 +0000 LLMs explained for backend engineers. The mental model you need before writing any AI code: stateless function, tokens, context window, sampling, cost. AI Engineering Your First LLM Integration: APIs, Errors, Retries https://aiengineerpath.dev/ai-engineering/articles/02-first-llm-integration.html https://aiengineerpath.dev/ai-engineering/articles/02-first-llm-integration.html Mon, 26 May 2026 09:00:00 +0000 Calling OpenAI and Anthropic APIs the right way. Streaming, rate limits, timeouts, retries, JSON parsing failures, idempotency. AI Engineering Prompting as Code: Structured Outputs and Templates https://aiengineerpath.dev/ai-engineering/articles/03-prompting-as-code.html https://aiengineerpath.dev/ai-engineering/articles/03-prompting-as-code.html Mon, 26 May 2026 09:00:00 +0000 Treat prompts like config. Version them, test them, force JSON schemas. Patterns for production prompt engineering. AI Engineering RAG Explained: How to Build an AI That Answers Questions From Your Documents https://aiengineerpath.dev/ai-engineering/articles/04-rag-system-design.html https://aiengineerpath.dev/ai-engineering/articles/04-rag-system-design.html Mon, 26 May 2026 09:00:00 +0000 The most common pattern for adding AI to your product, explained from scratch. Pipeline, components, and design choices, with diagrams and no jargon. AI Engineering Choosing Chunking Strategies: A Practical Framework https://aiengineerpath.dev/ai-engineering/articles/05-chunking-strategies.html https://aiengineerpath.dev/ai-engineering/articles/05-chunking-strategies.html Mon, 26 May 2026 09:00:00 +0000 The single highest-leverage decision in RAG. Four strategies and a decision tree to pick the right one for your documents. AI Engineering Hybrid Search: When Pure Vector Search Isn't Enough https://aiengineerpath.dev/ai-engineering/articles/06-hybrid-search.html https://aiengineerpath.dev/ai-engineering/articles/06-hybrid-search.html Mon, 26 May 2026 09:00:00 +0000 Why semantic search alone fails in production, and how to combine it with keyword search to get retrieval that actually works. AI Engineering Embeddings Deep Dive https://aiengineerpath.dev/ai-engineering/articles/07-embeddings-deep-dive.html https://aiengineerpath.dev/ai-engineering/articles/07-embeddings-deep-dive.html Mon, 26 May 2026 09:00:00 +0000 How embedding models work, when to swap them, and how to benchmark on your data. AI Engineering How to Know If Your AI Is Actually Working: A Beginner's Guide to Evals https://aiengineerpath.dev/ai-engineering/articles/08-llm-evaluation-pipeline.html https://aiengineerpath.dev/ai-engineering/articles/08-llm-evaluation-pipeline.html Mon, 26 May 2026 09:00:00 +0000 A friendly introduction to LLM evaluation. What evals are, why they matter, and how to build your first one. AI Engineering LLM-as-Judge: When It Breaks and How to Fix It https://aiengineerpath.dev/ai-engineering/articles/09-llm-as-judge.html https://aiengineerpath.dev/ai-engineering/articles/09-llm-as-judge.html Mon, 26 May 2026 09:00:00 +0000 Position bias, verbosity bias, self-preference, and how to write rubrics that survive contact with reality. AI Engineering Caching for LLM Apps https://aiengineerpath.dev/ai-engineering/articles/10-caching-for-llm-apps.html https://aiengineerpath.dev/ai-engineering/articles/10-caching-for-llm-apps.html Mon, 26 May 2026 09:00:00 +0000 Three layers of caching that cut LLM costs by 50 to 90% in production. Prompt caching, semantic caching, and embedding caching. AI Engineering Production AI Observability https://aiengineerpath.dev/ai-engineering/articles/11-production-ai-observability.html https://aiengineerpath.dev/ai-engineering/articles/11-production-ai-observability.html Mon, 26 May 2026 09:00:00 +0000 Why AI needs different observability than traditional services. Traces, evals in CI, and drift detection. AI Engineering Cost Optimization in Production AI https://aiengineerpath.dev/ai-engineering/articles/12-cost-optimization.html https://aiengineerpath.dev/ai-engineering/articles/12-cost-optimization.html Mon, 26 May 2026 09:00:00 +0000 Where AI money goes and how to spend 30% of what you do now. Model routing, prompt caching, eval-driven cost cuts. AI Engineering Build It: A Small RAG Service That Ships https://aiengineerpath.dev/ai-engineering/articles/13-capstone-build-rag-service.html https://aiengineerpath.dev/ai-engineering/articles/13-capstone-build-rag-service.html Tue, 26 May 2026 09:00:00 +0000 The capstone. Wire chunking, hybrid retrieval, structured generation, evals, caching, and observability into a working internal-docs Q&A service. The onboarding bot your team actually needs. AI Engineering What Is MCP? A Beginner's Guide to the Model Context Protocol https://aiengineerpath.dev/mcp-development/articles/01-what-is-mcp.html https://aiengineerpath.dev/mcp-development/articles/01-what-is-mcp.html Mon, 26 May 2026 09:00:00 +0000 MCP explained from scratch. What it is, why it exists, how the pieces fit, and how to use one in Claude Desktop or Cursor. MCP Development