The Biggest Flaw in My AI Evaluation Wasn't the Models. It Was My Scorecard.
I recently ran a small evaluation to compare three AI coding assistants. The task sounded straightforward: give each model the same engineering artifa…
Testing & QA
⚑ Report a ProblemLatest Testing & QA news from Tech News
All topics
agents
ai
api
architecture
automation
aws
backend
beginners
career
claude
cybersecurity
database
devchallenge
devops
javascript
llm
machinelearning
mcp
opensource
performance
productivity
programming
python
security
showdev
softwareengineering
testing
tutorial
typescript
webdev
A PR review agent with 2 comments beats one with 20
A good review agent needs a bias toward silence, a validation pass against its own findings, a context budget, and a definition of "correct" that come…
My AI coding agent reverted a core feature in one buried line — and its commit messages lied about it
2026-07-23 — Masih Moafi I build Elpis , a terminal coding agent forked from Codex-rs, focused on one thing: keeping your context window empty. Its wh…
Как я автоматизировал превращение вайбкодерского PoC в production-ready MVP
Как я автоматизировал превращение вайбкодерского PoC в production-ready MVP За несколько часов с помощью AI можно собрать работающий PoC: интерфейс от…
"My AI engineer has a second AI that tells it NO. Three times, it was right."
Yesterday afternoon I was about to ship a change to a production pipeline. The code was written, tested, deployed to the runtime directory. The nightl…
My Checklist for Reviewing AI-Generated Code
The agent handed me a function that fetched a Stripe customer, read customer.tax_ids.data[0].value , and used it as the VAT number for an invoice. Cle…
In an AI Builder + AI Reviewer World, What Should Developers Actually Review?
TL;DR If you frame yourself as the manager with final accountability, it becomes clearer where human review is necessary in an AI-driven workflow. You…
Where the Review Point Moved
Originally published at tddbuddy.com . Reviewing the diff is now harmful, not just insufficient. This is the first post in a two-part series, "The PR …
A Pipeline Where a Second AI Tries to Disprove the First One's Fix
Originally published at blog.whynext.app . Anyone who has had an AI agent do code review knows this. It's pretty good at other people's code. But have…
A Verdict Is Not Evidence. Test Is Where I Learned the Difference.
The call-order change came back pass-with-risk. I read the recommendation, saw it had a name and a reason, and felt the task close. Then I looked at t…
Why AI code review hallucinates — and the two gates that fix it
CCA-Audit — open source (MIT) AI code review has a trust problem, and it's not that it misses bugs. It's that it invents them. If you've run an LLM ov…
AI Code Review That Engineers Actually Trust: The Pipeline We Run on Every Pull Request
Bolting an LLM onto your pull requests is a weekend project. Building AI code review that your engineers don't disable within two weeks is the actual …
Don't take the model's word for it
A while back, a "read-only" code reviewer I was running rewrote my git history. Read-only, as it turns out, holds right up until the model decides a f…
The YC president open-sourced the stack he builds with. What it says about taste
Originally published on productize.life . Quick answer: gstack is an open-source (MIT) skill set that Garry Tan, president of Y Combinator, builds wit…
"When AI reviews AI's code, you've built an infinite loop. Here's how we stopped it."
While building an AI code review product (Orange Codens), there was one design problem that dominated everything else: when the reviewer is AI and the…
The PR Review Canvas – a free interactive checklist for better code reviews
Code review is one of the most valuable activities in software development — but also one of the most inconsistent. I've seen teams where every review…
Open the PR your reviewer has not met yet
I got a PR to review recently. Large diff, AI-assisted, touching a module three other features depended on. The description was one sentence. It named…
Working With AI: What Actually Works For Me
I think a lot of people still imagine AI coding as opening ChatGPT, asking for code, and copy-pasting the result. That's not really how I work anymore…
Stop Merging Blind: How I Use @code-reviewer Before Every PR
Think about the last time you merged a PR without a proper review. Maybe tests passed. Maybe it looked fine at a glance. Maybe you were moving fast an…
The AI-generated C# that passes review and breaks in production
TL;DR — AI assistants are producing C# that looks correct and passes review, but reintroduces production regressions we spent years training out of te…
LLM Smells: The Tells in AI Writing, and the Costlier Ones in AI Code
there is a piece by shrijal shrestha called "various llm smells" that put words to something i think a lot of us have been feeling but not naming. the…
What I learned building a tiny PR-review bot in 10 days
For the past ten days I've been building a GitHub bot that reviews pull requests against a team's own written conventions. Not a generic "linter with …
Multi-Line Formatting by Default
I have reviewed so much code that I usually prefer multi-line formatting by default. Not because it looks prettier. Because code review happens line b…
Human-on-the-Loop: AI Reviewing AI PRs at cortex (769 PRs/month, while raising the quality bar)
Hi, I'm Ryan , CTO at airCloset. Disclaimer : "cortex" in this article is the internal codename for an AI platform built in-house at airCloset. It is …
Agentic code review in production: orchestration, evaluation, and the cost of being wrong
What "agentic" actually buys you over a linter, why single-model approaches stall, and why false positives — not raw model capability — determine whet…
How I Use Claude for Code Review — Catching Bugs Before They Reach Production
Last month, our team's bug escape rate dropped from 23% to under 3% . We didn't hire more QA engineers. We didn't write more tests. We started using C…
Code Review Is Broken for AI-Generated Code — The Case for Pre-Implementation Governance
Your team's code review process was designed for a world where a developer wrote every line. In that world, a PR with 200 lines meant 200 lines of hum…