- 01
Evaluating LLM agents: building the design phase right
You built an agent and it works, sometimes. Visualisation, tracing and unit tests are what turn that into something you can change without fear.
Read note ->agents9 min - 02
From state to edges: how LangGraph connects the dots
States, nodes and edges are the whole of LangGraph. What each one is for, how reducers decide what survives an update, and a routing bot built out of all three.
Read note ->agents8 min - 03
RAG with LangChain, pgvector and OpenAI
Taking RAG past a mock lookup: building and storing real embeddings in Postgres, retrieving on similarity, and handing the context to GPT.
Read note ->rag6 min - 04
Understanding retrieval-augmented generation with OpenAI
A GPT model only knows what it was trained on. RAG is the pattern for handing it what it does not know, at the moment it needs it.
Read note ->rag5 min - 05
Passwordless authentication in a Cognito user pool
Cognito's custom auth flow is three Lambdas and a state machine you have to write yourself. What each trigger receives, what it must return, and where the tokens come from.
Read note ->auth7 min - 06
Converting a React app to native mobile with Capacitor.js
Capacitor wraps an existing React build as native Android and iOS apps. The ten steps from create-react-app to a project open in Xcode.
Read note ->mobile5 min