Notes, lessons and log lines from working with AI.https://peppesilletti.ioen-gbWed, 17 Jun 2026 00:00:00 GMThttps://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 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/ai-good-architecture-quality-codehttps://peppesilletti.io/ai/notes/ai-good-architecture-quality-codeAI doesn't change the fundamentals of good software — it amplifies whatever foundation you give it.Thu, 26 Feb 2026 00:00:00 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/building-software-products-with-aihttps://peppesilletti.io/ai/notes/building-software-products-with-aiMindset shifts and practical approaches for building software products with AI assistance.Thu, 26 Feb 2026 00:00:00 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/designing-agentic-workflowshttps://peppesilletti.io/ai/notes/designing-agentic-workflowsThe fundamentals of designing, evaluating, and debugging agentic AI workflows.Thu, 26 Feb 2026 00:00:00 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/designing-agents-as-teammates-not-toolshttps://peppesilletti.io/ai/notes/designing-agents-as-teammates-not-toolsEffective agentic AI requires treating agents as teammates who need proper context and experience design, not as tools you configure with the right parameters.Thu, 26 Feb 2026 00:00:00 GMTPeppeconcepthttps://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 GMTPeppelessonhttps://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 GMTPeppelessonhttps://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 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/planning-agentic-pattern-with-code-improves-performancehttps://peppesilletti.io/ai/notes/planning-agentic-pattern-with-code-improves-performanceLetting the LLM express its plan in software code that you can just execute can be a very powerful way to let it write rich plans.Mon, 19 Jan 2026 00:00:00 GMTPeppeloghttps://peppesilletti.io/ai/notes/agents-can-be-evaluated-objectively-or-subjectivelyhttps://peppesilletti.io/ai/notes/agents-can-be-evaluated-objectively-or-subjectivelyCan the evaluation condition be expressed mathematically? If so, we can evaluate with code. If it can't, it's a task that requires non-deterministic thinking.Wed, 14 Jan 2026 00:00:00 GMTPeppeconcepthttps://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 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/llm-as-a-judge-grading-with-a-rubric-gives-more-consistent-results-when-evaluating-reflection-agentic-patternhttps://peppesilletti.io/ai/notes/llm-as-a-judge-grading-with-a-rubric-gives-more-consistent-results-when-evaluating-reflection-agentic-pattern"LLM as a judge" can help by providing a rubric that the LLM can use in the reflection prompt to judge the output.Mon, 12 Jan 2026 00:00:00 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/reflection-agentic-pattern-can-use-external-feedback-from-tools-to-improve-outputhttps://peppesilletti.io/ai/notes/reflection-agentic-pattern-can-use-external-feedback-from-tools-to-improve-outputBy using a tool for web research, the LLM can adjust course with the external information.Mon, 12 Jan 2026 00:00:00 GMTPeppeloghttps://peppesilletti.io/ai/notes/agentic-design-patternshttps://peppesilletti.io/ai/notes/agentic-design-patternsReflection, tool use, planning and multi-agentic workflows.Wed, 07 Jan 2026 00:00:00 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/agentic-ai-can-be-less-or-more-autonomoushttps://peppesilletti.io/ai/notes/agentic-ai-can-be-less-or-more-autonomousFrom all steps predetermined to agents that make many decisions autonomously and can create new tools on the fly.Mon, 05 Jan 2026 00:00:00 GMTPeppeconcepthttps://peppesilletti.io/ai/notes/agentic-ai-works-better-for-well-know-linear-processeshttps://peppesilletti.io/ai/notes/agentic-ai-works-better-for-well-know-linear-processesAgents work better with a clear, step-by-step process and standard procedures to follow. Its job gets harder if the steps are not known ahead of time.Mon, 05 Jan 2026 00:00:00 GMTPeppeconcepthttps://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 GMTPeppeconcepthttps://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 GMTPeppelessonhttps://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 GMTPeppeconcept