Daily notes on AI, testing, and building software.
As AI-generated code becomes a default part of software development, a critical gap has emerged: the code LLMs write is largely untested in any systematic way. New research from ArXiv and a new open-source tool called…
As LLM-based agents become production software components, QA teams face a new challenge: traditional test automation was designed for deterministic systems, but AI agents are non-deterministic by nature. New research…
OpenAI's expanded Codex plugin ecosystem — now including CircleCI, CodeRabbit, GitLab Issues, and Atlassian Rovo — transforms Codex from a code generation tool into a full CI/CD-aware testing collaborator capable of…
OpenAI's GPT-5.3-Codex marks the shift from AI that helps you write tests to AI that runs, debugs, and iterates on them autonomously for hours at a time — and it did all of this while helping build itself. For QA teams,…
Google just open-sourced research on an LLM-powered tool that correctly diagnoses the root cause of integration test failures 90% of the time — at massive scale. If this approach becomes standard, it could eliminate one…
Google's newly published Auto-Diagnose system achieves 90% root-cause accuracy on integration test failures using LLMs — tackling one of the most time-consuming, cognitively draining parts of QA. If your team spends…
CVE-2026-35616 is a critical pre-authentication API access bypass vulnerability in Fortinet FortiClient Endpoint Management Server (EMS), scored CVSS 9.1. An unauthenticated attacker with network access to the EMS…
CVE-2026-34197 is a remote code execution vulnerability in Apache ActiveMQ Classic that exploits the broker's built-in Jolokia HTTP-to-JMX bridge, allowing an attacker to force the broker to fetch and execute a remote…
CVE-2026-33824, nicknamed BlueHammer, is a zero-click, unauthenticated remote code execution vulnerability in the Windows Internet Key Exchange (IKE) service extension (IKEEXT.dll) with a maximum-severity CVSS score of…
CVE-2026-33017 is a critical unauthenticated remote code execution (RCE) vulnerability in Langflow, the popular open-source framework used for building AI agent and RAG pipelines. The flaw allows a remote,…