Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: evals
Building an Agent Improvement Loop with LangGraph and LangSmith Evals
A step-by-step LangGraph and LangSmith loop: trace collection, three scoring layers, failure clustering, offline regression datasets and pre-ship evals.
24 min readVibe Coding vs Agentic Engineering: What Google's New SDLC Paper Argues
A summary of Google's whitepaper on the new SDLC: verification as the dividing line, harness over model, context engineering, the 80% trap and conductor versus orchestrator.
9 min readEvals for GenAI and Agents: Single-Turn vs Multi-Turn Testing in CI
Compare trace-based single-turn evals with simulator-driven multi-turn evals, covering goldens, LLM judges, agent metrics, CI gates and seven working practices.
8 min readEvaluating AI Agents: Golden Sets, LLM Judges and CI Regression Gates
Offline and online evaluation for agents with AgentCore, LangFuse, Ragas and DeepEval, plus golden datasets, CI gating, A/B tests and memory feedback.
18 min readAuditing Order Confirmations an Agent Never Placed: Checking Values Instead of Wording
How to build a claim-consistency evaluator for a support agent that places real orders: partition outcomes by owner, verify identifiers against tool results.
16 min readKeeping a Vector Store Honest: Lifecycle, Tenancy, Metadata, Hybrid Search
Running a vector store for RAG in production: document lifecycle, embedding refresh, tenant filters, hybrid search for exact IDs, metadata and tracing.
9 min readEval-Driven Development for AI Agents: Contracts, Traces and Judges
Treat evals as CI/CD for agents: version prompts as contracts, trace runs with OpenTelemetry and Tempo, then test with deterministic checks and an LLM judge.
13 min readAgent Evals With a Travel Bot: Outcome vs Trajectory, Gates and pass^k
A worked travel-booking example of evaluating agents: outcome and trajectory axes, deterministic gates before weighted scores, a 2x2 matrix and pass@k vs pass^k.
8 min readAgent Skills Are Code: Scope, Disclose, Script and Test Them
Why SKILL.md is production logic: write procedural knowledge only, tune activation, load progressively, move exact work to scripts, build evals and delete losers.
10 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.