Articles for people who
ship the stack
Original rewrites on React, Node.js, TypeScript and AI — practical notes from the same engineering practice behind our operator software. Article bodies are in English.
Tagged: Testing
Who Grades the Grader? Validating Agent Eval Suites Before You Trust Them
Who validates the eval suite? Judge drift, outcome-only scoring, private failure modes and hidden cost can make LLM evals lie. Checks for the suite itself.
7 min readWhy Agents Pass Every Offline Eval and Still Fail in Production
Six drift modes quietly age an eval set. A four-axis trace score and a weekly cluster-and-promote loop keep CI honest against what live traffic actually does.
9 min readWhat Agentic Testing Changes for QA and Where It Still Falls Short
Scripted and self-healing suites break when the UI shifts. Intent-driven browser agents cut maintenance, but regulated flows and odd UIs still need care.
7 min readEngineering an Evaluation Control Plane for Production AI Agents
How to build continuous agent evaluation: separate execution and control planes, trace-based assertions, golden datasets, CI gates, lifecycle testing and a launch checklist.
10 min readPlaywright or Browser Agents? Where Scripted Automation Still Wins
Playwright scripts vs MCP-driven browser agents: brittleness, determinism and cost, one invoice task done both ways, the security risks and when each wins.
8 min readFront-End Browser Testing That Holds Up: Journeys, Assertions and Flaky Tests
A practical approach to front-end testing: choose critical user journeys, write outcome assertions, control staging data, cut flakiness and review AI-generated Playwright tests.
8 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 and Skill Evals in Practice: Trajectories, Graders, pass^k and Tools
How to evaluate agents and skills: trigger and execution suites, grader strength order, precision and recall, pass@k vs pass^k, a tiny Python harness and tool picks.
11 min readSemantic Retrieval with ChromaDB and Embeddings for a Multi-Agent Platform
Moving an enterprise multi-agent platform to real semantic retrieval: word-aware chunking, cosine distance in ChromaDB, test layers and the next RetrievalAgent step.
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.