Verifying Agentic Development at Scale (via @ido_pesok)kept by eddie • jun 01Cognition built end-to-end testing inside Devin's virtual machine to verify agentic software development at scale.AboutCognition · DevinFiled#ai-coding#automation#developer-tools#software-engineering#testingRelatedHow Cognition Uses Devin to Build Devin - by Nader DabitCognition uses its own Devin agents across Slack, Linear, and CLI to merge 659 PRs weekly, automating code review, bug triage, and design-system audits.also on Cognition, Devin, #automation, #ai-coding, #software-engineering, #developer-toolsIntroducing FrontierCodeCognition's FrontierCode benchmark measures code mergeability, finding even top models like Claude Opus 4.8 score only 13.4% on its hardest 50 tasks.also on Cognition, Devin, #ai-coding, #software-engineering, #developer-toolsThe Evolution of the Agent Harness - by Dan McAteerThe article argues that as AI models internalize agent harness capabilities, the remaining harness evolves into an interface for managing scarce human attention rather than directing model behavior.also on Devin, #automation, #ai-coding, #software-engineering, #developer-toolsIntroducing the Safari MCP server for web developersApple released a Model Context Protocol server for Safari that lets AI agents inspect DOM, network requests, screenshots, and console output to debug websites autonomously.also on #testing, #ai-coding, #software-engineering, #developer-toolsDon't rely on instructions, use Agent Hooks to enforce guardrails – Zarar's blogAgent hooks (PreToolUse and Stop) enforce coding rules deterministically, preventing raw HTML tags and blocking task completion until a design-system test passes.also on #testing, #ai-coding, #software-engineering, #developer-tools
Verifying Agentic Development at Scale (via @ido_pesok)