I build the systems that work while you don't.

On AI, sound money, and owning your own life.

Twenty years of the same idea, across three domains.

I build income with AI, store value in crypto, reason through economics, and started it all with a story about dinosaurs.

Last updated: 2026-06-16. Public GitHub now has 24 repos. The Big 6 is the WIP direction; the shipped/high-signal repos below are the current proof.

Most AI demos look great until they meet real users, messy data, or production constraints. I build the opposite.

This isn't cost arbitrage. It's engineering arbitrage.

Systems that work in production. Not demos that break on real data.

Production AI is the execution surface: RAG pipelines with grounded retrieval, evaluation flows that catch regressions, agent infrastructure that does not trust the LLM, and observability so you know what is failing and why.

But the identity is not a narrow AI title. The portfolio is a map of the worldview: economics, AI, data, crypto, finances, and entrepreneurship as one system.

Three years building production AI systems inside a real business — not side projects. Internal assistants, content-generation pipelines, scraping workflows, RAG systems, business automation. The result: 90% operational cost reduction, 50% faster content production.

The interesting part: Expat Money helps people structure their lives across jurisdictions — flag theory, expat planning, financial sovereignty. The AI wasn't decoration. It was how a business built around individual autonomy automated the labor of producing expert knowledge at scale. That's what grounded retrieval looks like in a domain where wrong answers have real consequences.

Six systems, one per field across the three domains and the spine. They are build-in-public concepts until their repos exist and contain code.

• catallaxy — Economics — The Calculation Lab — proves the knowledge problem is computable.
• understudy — AI — The Glass Box — proves AI that runs without you, not against you.
• groundwork — Data — Corpus Autopsy — proves clean local data precedes grounded AI.
• proof-of-cost — Crypto — Real-Yield Calculator → Verdict Gallery — proves mechanism over narrative.
• sound-money — Finances — The Time Machine — proves the unit of account is the variable.
• operator-os — Entrepreneurship — The Operating Console — proves a one-person business is a system.

These are the existing repos I would inspect first.

• WCP-Compliance-Agent-V5 — Production-ready WH-347 compliance platform: deterministic validation decides, the LLM explains, and every decision is traceable.
• aria-agent — Controlled AI agent framework and conceptual ancestor of understudy.
• groundtruth — Grounded document QA / RAG proof with citations and approval workflows.
• llm-gateway — LLM proxy with routing, guardrails, cost control, and fallback.
• agenttrace — Agent observability and replay SDK with cost attribution.
• evalforge — Regression testing framework for RAG and agentic AI.
• github-issue-pr-agent — Autonomous issue-to-PR workflow with a safety boundary.
• knowledgeops — Unified AI operations platform with microservice architecture.

All 24 public repos are listed on the portfolio, each with its own page:

fishraposo.github.io/projects.html

The catalog separates:

• Big 6 WIP direction
• highlighted shipped/relevant systems
• supporting infrastructure
• meta/profile repos
• archive/study repos

Send me your broken AI workflow. I'll tell you what's breaking, why it's breaking, and the fastest path to a system your team can actually depend on.

Canonical surfaces: GitHub profile, portfolio site, and the positioning source in career/Main Docs/.