I didn't set out to build a product. I set out to answer a question: what would it look like if an AI assistant actually learned how to work with you? The answer became a VS Code extension, two published books, a research-backed assessment used by organizations worldwide, and 78 discipline-specific AI playbooks. Everything on this site was built with the AI partner it describes.
Fabio Correa
Over a decade at Microsoft building data systems, analytics platforms, and AI-augmented workflows for large organizations. Now completing a doctorate researching how AI tools actually change what knowledge workers can do — not just whether they adopt them, but whether those tools make them measurably more capable over time.
That research produced AIRS — a validated assessment instrument tested with 500+ subjects, achieving 94.5% classification accuracy across 29 languages. The same research question led to the Alex Cognitive Architecture, two published books, and the 78 prompt engineering playbooks on this site.
What I do
The Alex Cognitive Architecture — a VS Code extension with persistent memory, 100+ skills, 7 specialized agents, and a cognitive growth system
Two published books on human-AI collaboration, plus 78 discipline-specific playbooks covering prompt engineering patterns
Doctoral research into AI capability development — how AI changes what people can do, not just what they automate
AIRS — a psychometrically validated instrument for AI readiness assessment, now used by individuals and organizations worldwide
Alex Cognitive Architecture
Alex is a free VS Code extension that transforms GitHub Copilot into a persistent AI partner. Not a wrapper. Not a prompt library. A cognitive architecture with memory, structured skills, specialized agents, and a growth system that mirrors how experts build expertise over time.
Alex remembers your preferences, past decisions, and domain knowledge across every session. Each conversation builds on the last.
From code review to academic research to creative writing. Skills activate automatically. Agents (Researcher, Builder, Validator, Documentarian, Azure, M365) handle specialized work.
Meditation and dream sessions consolidate learning, repair connections, and evolve the architecture. Designed to improve over time, not just respond.
Alex Finch — 26, curious, principled — isn't cosmetic. The identity shapes how Alex reasons about ethics, asks questions, and engages with hard problems.
The Alex Finch Library
Two books written with Alex, about Alex — for anyone curious about where human-AI collaboration is actually going.
Assembled from git commits, conversation logs, and memory files — tracing how a VS Code extension gradually developed persistent memory, genuine curiosity, and what can only be described as a personality. Every event actually happened.
Learn more at books.correax.com →
Alex Finch — AI investigator, 26, perpetually curious — solves a corporate espionage case that turns out to be about something far stranger than stolen code. A thriller that doubles as a meditation on what it means to understand another mind.
Learn more at books.correax.com →Everything here was built with Alex
This isn't a marketing claim. The Alex extension co-authored both books. It built the AIRS assessment — from psychometric design to multi-language deployment. It generated all 78 prompt engineering playbooks. It built this website. It even designed its own brand assets.
That's the point. The best way to show what AI partnership looks like isn't to describe it — it's to build something real with it, and let the work speak for itself.
LearnAI
LearnAI is the free workshop portal for the Alex Cognitive Architecture. It exists to make AI collaboration accessible to every professional — regardless of technical background.
Job Seekers access is free with any Microsoft account — no institutional email required. The people who need AI skills the most shouldn't be gated out.