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: agent-evaluation
Self-Improving Agents: Healing, Learning and Controlled Change
Self-improving agents without the danger: runtime self-healing, learning across runs, and changes that ship only after evals and canaries, on LangGraph.
15 min readAgent Evaluation and Optimisation: DSPy, TextGrad, Reflexion and OPRO Compared
How DSPy, TextGrad, Reflexion and OPRO fit together as feedback loops for agents, which to adopt first, and how to separate evaluation from optimisation in an enterprise.
12 min readA Minimum Viable Experimentation Platform for AI Agents: A/B Tests and Shadow Runs
How to decide whether agent version B should replace A, using pipeline definitions, local assignment, exposure events, traces, sandbox evaluation, shadow runs and guardrails.
11 min readAI Agent Equals AI Model Plus Harness: What the Harness Does and Why
An agent is a model plus a harness. See how harnesses add tools, memory, loops and guardrails, how open-weight and open-source agents differ, and why evals matter.
10 min readA Stable Agent Name Is Not the Artifact You Reviewed: Pin the Candidate
An agent's name can stay the same while its skills, tools and permissions change. Pin name, behavior candidate and run identity apart for review and rollback.
8 min readThe Free Verifier Problem: Why Agent Loops Stall Once the Work Isn't Code
Coding agent loops work because compilers and tests verify for free. Outside code there's no free verifier: four subsidies, failure data and an oracle test.
6 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.