Blog
Notes on Building Production AI
Practical write-ups on LLMOps, RAG, agentic AI, governance, and lessons from building enterprise AI platforms.
AI Coding Agents in the Enterprise: Review Gates, Security Boundaries, and What to Measure
A field guide to adopting AI coding agents in an enterprise engineering organization — autonomy tiers, review gates, security boundaries, and the metrics that distinguish real velocity from quality debt.
Why Most AI Agents Break in Production — And What Reliable Agents Actually Look Like
The failure modes that separate demo-grade agents from production-grade ones — and what production-grade agents actually require beneath the surface.
Why Most Enterprise RAG Projects Stall at 70% Accuracy — And What Actually Fixes It
The diagnostic mistakes that keep RAG systems plateaued at demo quality, and the changes that move them into production reliability.
How to Build a Production RAG Pipeline — A Technical Walkthrough
A component-by-component guide to assembling a RAG pipeline that holds up under real production load — ingestion, chunking, embeddings, retrieval, reranking, generation, and the eval harness around it.
How AI Systems Actually Fail — And How to Test Them for Security
Why most AI security approaches fail in production — and how to test systems the way they actually behave.
The AI Governance Questions Every Board Should Be Asking
A practical set of questions boards and executives should be asking about their AI programs — covering risk, compliance, accountability, and operational readiness.
From Prototype to Platform: A Practical View of LLMOps
Key steps teams often miss when moving from a successful AI demo to a reliable, governed production system.
Get new posts delivered to your inbox
Subscribe on Substack for articles on LLMOps, RAG, agentic AI, and production AI engineering.