Peppe’s Garden — AI notesNotes, lessons and log lines from working with AI.https://peppesilletti.ioen-gbWed, 17 Jun 2026 00:00:00 GMTCoding Patterns Stop AI Forgetfulnesshttps://peppesilletti.io/ai/notes/coding-patterns-stop-ai-forgetfulnesshttps://peppesilletti.io/ai/notes/coding-patterns-stop-ai-forgetfulnessUsing a strict coding pattern helps an AI write better code by giving its limited short-term memory a clear blueprint to follow.Wed, 17 Jun 2026 00:00:00 GMTPeppeSoftware engineers won't need to write and read code only when agents will embed good engineering practices in their DNAhttps://peppesilletti.io/ai/notes/software-engineers-wont-need-to-write-and-read-code-only-when-agents-will-embed-good-engineering-practices-in-their-dnahttps://peppesilletti.io/ai/notes/software-engineers-wont-need-to-write-and-read-code-only-when-agents-will-embed-good-engineering-practices-in-their-dnaAs long as we have to tweak the code ourselves, I don't think much will change. In fact, good engineering practices help AI generate better solutions.Thu, 12 Feb 2026 00:00:00 GMTPeppeThere's not difference in maintainability in AI vs Human generated codehttps://peppesilletti.io/ai/notes/theres-not-difference-in-maintainability-in-ai-vs-human-generated-codehttps://peppesilletti.io/ai/notes/theres-not-difference-in-maintainability-in-ai-vs-human-generated-codeLLMs are great at pattern recognition, so as long as the codebase is well designed and architected, it'll produce high-quality output.Thu, 12 Feb 2026 00:00:00 GMTPeppeWe should invest in AX (Agents Experience)https://peppesilletti.io/ai/notes/we-should-invest-in-ax-agents-experiencehttps://peppesilletti.io/ai/notes/we-should-invest-in-ax-agents-experienceAI agents are becoming our new teammates: we should treat them as a new entity that requires its own UX to work properly, a bit like DX.Mon, 09 Feb 2026 00:00:00 GMTPeppeAgentic workflows are only as good as their weakest component + error analysishttps://peppesilletti.io/ai/notes/agentic-workflows-are-only-as-good-as-their-weakest-component-error-analysishttps://peppesilletti.io/ai/notes/agentic-workflows-are-only-as-good-as-their-weakest-component-error-analysisAgentic workflows follow the same principle as Theory of Constraints: look at the traces, count the errors, fix the bottleneck.Wed, 14 Jan 2026 00:00:00 GMTPeppeSoftware systems integrating with AI still benefits from good software architecturehttps://peppesilletti.io/ai/notes/software-systems-integrating-with-ai-still-benefits-from-good-software-architecturehttps://peppesilletti.io/ai/notes/software-systems-integrating-with-ai-still-benefits-from-good-software-architectureAI is infrastructure, not architecture. The good practice is still to create an abstraction exposing a stable API.Mon, 05 Jan 2026 00:00:00 GMTPeppeCoding with LLMs makes me think morehttps://peppesilletti.io/ai/notes/coding-with-llms-makes-me-think-morehttps://peppesilletti.io/ai/notes/coding-with-llms-makes-me-think-moreWhen working with AI-assisted tools like Cursor, I'm forced to think more about what I want the agent to do, so that it doesn't get lost.Sat, 03 Jan 2026 00:00:00 GMTPeppeContext compression for better code generationhttps://peppesilletti.io/ai/notes/context-compression-for-better-code-generationhttps://peppesilletti.io/ai/notes/context-compression-for-better-code-generationResearch, planning, implementation: an iterative process aimed to improve the AI context.Fri, 02 Jan 2026 00:00:00 GMTPeppe