{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Slightly Caffeinated","title":"Fall Vibes, AI Evals, and Code Review Bots","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/9b0bebe3\"></iframe>","width":"100%","height":180,"duration":1678,"description":"TJ Miller and Chris Gmyr compare the AI code review bots they are each building at work, from fixture repos of real PRs to requiring a line of code behind every finding. TJ also shares the eval-driven wizard he built to turn source material into learning content, sharing insights on building AI systems you can test and trust.\nLinks\nAsk a question or suggest a topic\nPodcast: Site | Twitter/X | Bluesky\nChris: Site | Twitter/X | Bluesky\nTJ: Site | Twitter/X | Bluesky\nReferences Mentioned\nHive - Chris's orchestration framework\nGorillaz - the show TJ saw in Detroit\nDeltron 3030 - the opener and TJ's all-time favorite hip hop artist\nLuma Brighter Learning - where TJ is building the learning content wizard and the PR review bot\nClaude Opus 5.5 - the model TJ wants to move the content wizard to, then rerun the evals\nPrism - TJ's Laravel AI package, which needs updates before the model upgrade\nGitHub Copilot - the reviewer that caught real issues TJ's bot missed, and the inline comment style Chris wants to copy\nCodeRabbit - one of the review services both teams passed on in favor of building their own\nAmazon Bedrock - where Chris's reviewer runs to keep code inside their own infrastructure","thumbnail_url":"https://img.transistorcdn.com/R-444p4eT9rQuQTWqDaqd1s9ThBWdFeMfb3HoJxPWnQ/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS8yODA0/N2JjZTg1ODNkZTQ4/MGM4N2M3NDgzZjQz/MzA5MS5wbmc.webp","thumbnail_width":300,"thumbnail_height":300}