Prompt Engineering in Production Is a Systems Problem
The prompt is the smallest part of the system. What determines output quality is the context, the constraints, the feedback loop and the evaluation set.
All posts
14 articles, newest first. If you know what job you need done, the three columns below are a faster route.
AI Code Assistants: Cursor, Claude Code and Copilot in Production
Six months of AI coding assistants on a real .NET and Python codebase: where they genuinely help, where they generate noise, and what they cost.
Write GEO Content That Gets Cited by Answer Engines
Structure pages so generative engines can lift them — a 40-60 word direct answer, extractable facts, matching structured data, and crawler access.
LLM Router vs LiteLLM Proxy vs Hosted Gateways: A Practical Guide
Three ways to route across models, compared on control, fallback behaviour, cost control, observability and portability — including when you need none of them.
Build Your First RAG Pipeline That Doesn't Hallucinate
A working retrieval-augmented pipeline with chunking, reranking and an evaluation harness — plus the three failure modes that produce confident wrong answers.
Embedding Model Review: Open-Source vs Hosted for RAG
bge-m3, gte, nomic-embed against hosted APIs: multilingual reach, long text, dimension cost, retrieval quality and the index rebuild trap.
How to Secure an AI Agent That Can Call Tools
Harden an agent that reads the web, calls APIs and writes files — where injected instructions land, plus least privilege, approval gates and audit logs.
Pinecone vs Qdrant vs Weaviate: Vector Database Comparison
Three vector databases compared on deployment, filtering, latency, ops burden, cost model and scaling — plus which one fits your team.
How to Evaluate an LLM Pipeline So You Can Change It Safely
Build an evaluation set from real production traffic, assert on properties instead of exact strings, and wire it into CI so prompt changes ship on evidence.
n8n vs Make vs Zapier: Workflow Automation for AI Workloads
Three automation platforms compared on self-hosting, code escape hatches, task billing, retries and long-flow reliability — with a pick for each team size.