My thoughts on software development, life, and more. See every page.
The AI Engineering Core Stack
The pieces that show up in every real AI product: model, orchestration, retrieval, app backend, frontend, and monitoring.
2026-08-20
Evals and Observability
If you cannot evaluate quality, you cannot improve it. Trace prompts, context, outputs, cost, and failures.
LLMs and Prompting
Context windows, hallucinations, system prompts, and why good prompting is mostly reducing ambiguity.
RAG and Knowledge Systems
Chunking, embeddings, ranking, and why search quality is often the real bottleneck — not the model. Includes RAG vs fine-tuning and how I implement it.
Shipping AI to Production
Caching, fallbacks, rate limits, prompt versioning, and cost control — how AI systems stay reliable after the demo.
Want to call me? book a call and I'll respond in the available time slot. I will ignore all soliciting.