https://alexmlewis.comhttps://alexmlewis.com/abouthttps://alexmlewis.com/contacthttps://alexmlewis.com/noteshttps://alexmlewis.com/notes/how-this-site-is-builthttps://alexmlewis.com/resumehttps://alexmlewis.com/start-herehttps://alexmlewis.com/workhttps://alexmlewis.com/work/ai-workflow-kithttps://alexmlewis.com/work/long-running-coding-agentshttps://alexmlewis.com/work/mcp-workflow-infrastructurehttps://alexmlewis.com/work/probabilistic-prompt-selectionhttps://alexmlewis.com/writinghttps://alexmlewis.com/writing/50-ai-skills-for-the-parts-of-your-day-you-never-get-around-tohttps://alexmlewis.com/writing/a-guide-to-setting-up-vercel-log-storage-in-aws-s3https://alexmlewis.com/writing/automating-meeting-note-management-with-krisp-zapier-google-drive-and-notionhttps://alexmlewis.com/writing/building-the-selector-retrieval-bandits-and-cold-start-solutionshttps://alexmlewis.com/writing/claude-code-the-hidden-testing-ground-for-ai-agentshttps://alexmlewis.com/writing/creating-a-blog-with-gatsbyjs-part-onehttps://alexmlewis.com/writing/creating-a-blog-with-gatsbyjs-part-twohttps://alexmlewis.com/writing/enhancing-retrieval-augmented-generation-with-xml-promptinghttps://alexmlewis.com/writing/ground-truthing-llm-judgeshttps://alexmlewis.com/writing/mastering-conditional-types-in-typescripthttps://alexmlewis.com/writing/one-thing-balancing-hard-and-soft-skills-in-the-age-of-aihttps://alexmlewis.com/writing/one-thing-cursor-claude-and-the-mocking-strugglehttps://alexmlewis.com/writing/one-thing-kneading-a-new-hobbyhttps://alexmlewis.com/writing/reward-engineering-and-evaluation-making-your-system-learn-what-mattershttps://alexmlewis.com/writing/staying-sane-on-a-6-000-line-refactor-with-claude-codehttps://alexmlewis.com/writing/understanding-useeffect-vs-uselayouteffect-in-react-a-comprehensive-guidehttps://alexmlewis.com/writing/use-insights-to-improve-your-claude-code-setuphttps://alexmlewis.com/writing/why-returning-cleanup-functions-in-useeffect-matters-in-reacthttps://alexmlewis.com/writing/why-static-prompts-fail-and-how-probabilistic-selection-solves-real-problems