Guides

Standalone decision guides and comparisons: the questions you hit while shipping LLM systems, answered with tables, trade-offs, and runnable code. For the structured curriculum, start with the AI Engineering track.

• LLM API Retries, Timeouts, and Fallbacks: A Production Playbook

  An error-by-error playbook for LLM calls: which failures to retry (and which never), stage-aware timeouts, model fallback chains, and why naive retries are a money bug.

  Last verified July 13, 2026 · 6 min read

• pgvector vs Pinecone vs Qdrant: Choosing a Vector Store for Your Existing Backend (2026)

  A backend engineer's decision framework for vector storage: when pgvector in your existing Postgres wins, when a dedicated engine earns its keep, and the sync-pipeline tax nobody prices in.

  Last verified July 11, 2026 · 6 min read

• How to Stream LLM Responses: SSE vs WebSockets (and the Fetch-Stream Option)

  Which transport to stream LLM tokens through your backend (SSE, raw fetch streaming, or WebSockets), with a relay implementation, proxy gotchas, and a decision table.

  Last verified July 13, 2026 · 5 min read

• How to Add an LLM to Your Existing Backend Without Rewriting It: 6 Integration Patterns

  Six production patterns for adding LLM features to an existing backend, with code: inline endpoint, queue worker, internal gateway, pgvector retrieval, event enrichment, and flagged rollout.

  Last verified July 13, 2026 · 10 min read