AI Engineering

Building products on top of language models: evals, retrieval, prompt design, latency and cost in production.

3 posts

LLM28 Sept 2026· 1 min read

Evaluating LLM output without vibes

Build a small eval harness before you touch the prompt.

#llm
#evals
#testing

RAG24 Sept 2026· 1 min read

RAG basics: chunking and retrieval

Why most RAG failures are retrieval failures.

#rag
#embeddings

Performance18 Sept 2026· 1 min read

Prompt caching and latency

Cut cost and time-to-first-token by reusing the prefix.

#llm
#caching
#latency