Head of DevOps & IT, Ionix.
Explore with your AI assistant
ChatGPT (opens in a new tab)Claude (opens in a new tab)Perplexity (opens in a new tab)Grok (opens in a new tab)
Opens a new chat with a prompt that points the assistant to this archive. What it writes is its own, not the archive's: check any quotation on its passage page before you rely on it. Or add the archive to your assistant as a connector: https://the-ai-record.com/mcp (how).
Head of DevOps & IT at Ionix, per the byline of the company's August 2026 blog post on its AI pull request reviewer.
Why their word counts: Wrote the post about a system his team built and runs; an insider describing his own company's tool, with no figures to audit.
Profile from this source, checked 2026-11-08.
An agent that has silently lost its tools does not report that it lost its tools. It answers anyway, with the same fluency as before.
Arthur VinbergHead of DevOps & IT · Leader, 
Ionix
Context. From the post's account of a sandbox failure in which the review agent could not run any commands.
Post: We built an AI PR reviewer. The hard part was teaching it to say nothing., Ionix blog, August 13, 2026, at Two failures that taught us more than the design did.
The first was cost. We were paying more for it than the results justified, and the obvious replacements wanted the same kind of money.
Arthur VinbergHead of DevOps & IT · Leader, Ionix
Context. The post's author explains why Ionix dropped Rovo, an AI review tool, before building its own reviewer; “it” in the passage is Rovo.
Post: We built an AI PR reviewer. The hard part was teaching it to say nothing., Ionix blog, August 13, 2026, at Why we built it instead of buying it.
Most days the entire output is one sentence saying the change looks fine, which is what we built it to do. On the days it leaves an inline comment, someone reads it.
Arthur VinbergHead of DevOps & IT · Leader, Ionix
Post: We built an AI PR reviewer. The hard part was teaching it to say nothing., Ionix blog, August 13, 2026, at If you’re building one of these.
The second is permission to find nothing. The prompt says outright that returning zero findings is the correct and expected outcome for a clean PR, and that findings must never be manufactured to look thorough.
Arthur VinbergHead of DevOps & IT · Leader, Ionix
Post: We built an AI PR reviewer. The hard part was teaching it to say nothing., Ionix blog, August 13, 2026, at The noise budget.
So the reviewer is deliberately a different stack from our authoring tools: OpenAI’s Codex CLI, pointed at a model deployment we host on Azure AI Foundry.
Arthur VinbergHead of DevOps & IT · Leader, Ionix
Post: We built an AI PR reviewer. The hard part was teaching it to say nothing., Ionix blog, August 13, 2026, at A reviewer that didn’t write the code.
A lot of our code is now written with an AI assistant in the loop. The first thing we tried was the obvious thing: ask the same Claude session that wrote the change to review it. That does not work. It missed critical bugs, and it missed them confidently, because it had already reasoned its way to believing the code was correct, and re-reading its own work only confirmed the conclusion.
Arthur VinbergHead of DevOps & IT · Leader, Ionix
Post: We built an AI PR reviewer. The hard part was teaching it to say nothing., Ionix blog, August 13, 2026, at A reviewer that didn’t write the code.