~/blog · ls -la

thinking
out loud.

Field notes on AI-first development, context preservation, and the strange new craft of orchestrating agents instead of typing code. Written from València, broadcast worldwide, occasionally argumentative.

$ rss available · no tracking · discussion lives on github
category / framework
framework.
07 entries · how to think about ai-first.
category / security
security.
02 entries · shipping safely with agents.
category / productivity
productivity.
01 entries · real gains, not theatre.
[blog/02] · entries — reverse chronological $ sort -r < index.log
2026-06-07 framework Spec-driven development: a green build is not a correct feature The same agent, same prompt, ships correct code on one run and a broken build-green diff on the next. I measured how often, and what actually makes the result something you can trust. #paelladoc#spec-driven-development#ai-coding#verification 7 min ./read → 2026-06-07 product Making product means installing a behavior, not running a feature factory The Bloomberg Terminal is ugly, looks like the 80s, and costs $32,000 a year. Nobody leaves it. The reason isn't the price, and it's everything about the difference between making product and shipping features. #paelladoc#product-management#product-strategy#feature-factory 7 min ./read → 2026-06-07 framework Locally correct, globally incoherent: the real problem with AI-generated code AI doesn't write incoherent code because it's dumb. It writes it because nobody manages what it sees each time it decides. The problem isn't inside the model, it's the missing layer. #paelladoc#ai-coding#software-architecture#context-engineering 5 min ./read → 2025-05-04 framework Stop guessing: 5 AI architecture patterns that actually prevent project failure Battle-tested architectural solutions for AI systems that last beyond the prototype stage #paelladoc#ai-architectures#ai-first-development 10 min ./read → 2025-04-27 framework The AI-First Development Framework Guide Discover the AI-First Development Framework, a methodology that preserves context and intent alongside code, ensuring sustainable, maintainable software in the... #ai-first-development#context-preservation#intent-driven-architecture 16 min ./read → 2025-04-17 security Securing AI code with Snyk: A practical guide Essential security strategies for protecting your AI-first development against hidden vulnerabilities #ml-pipeline-security#snyk#ai-first-development 13 min ./read →
— end of log · 6 entries · more cooking —