Building a RAG service: architecture, trade-offs, and lessons
·8 mins
Built a production-shaped RAG v1: offline indexing + FastAPI /query on Cloud Run (Terraform), with sources, logs, and basic monitoring.
Deep dives on platform engineering and AI systems: retrieval, evals, observability, and cost.
Built a production-shaped RAG v1: offline indexing + FastAPI /query on Cloud Run (Terraform), with sources, logs, and basic monitoring.