Forja WODs
Daily workouts, session log and a 1RM percentage calculator. Works without an API key; AI is optional.
Facundo Matias Aquino · applied AI
I build real software by directing AI agents.
I love creating with artificial intelligence. I make made-to-measure tools —and the odd game— by directing a team of AI agents: I bring the idea and the judgement; they write the code. From the clinical lab I bring the rigour: test, verify and document.
EU citizen · open to opportunities across Europe, especially Switzerland and Italy
01 / projects
Made to measure.
Tools I use every day and a game I wanted to play. Everything can be opened, downloaded or played.
Healthy Pockets
Track and project your investments in one place, 100% local. AI agents propose; the user decides.
Screenshots use fictional demo data. It does not connect to brokers or execute trades.
First 100 Days
Educational pixel-art simulator: you take office as president and have 100 days to make 7 decisions, each with its benefit and its cost.
A demo, not the full game. Simplified model: not a forecast or real policy. No AI at runtime.
“Business logic is going to move to these agents.”
My take: software stops being the same for everyone and starts adapting to each person and context. That is what I have been doing: tools made to measure.
02 / why I build
I always wanted to program. What I never liked was writing code.
AI resolved that contradiction. Now each idea has its own origin:
FORJA WODS · MADE TO MEASURE
There were workout apps, but none the way I wanted. So I made my own.
HEALTHY POCKETS · A REAL NEED
I wanted to see my finances clearly, with something more alive than a spreadsheet.
FIRST 100 DAYS · CURIOSITY
I watched two videos, an idea came up and I wanted to see if I could turn it into a game.
01Verify before trustingWhat the lab taught me: tests, limits and documentation come before a flashy demo.
02Simple to useIt has to work for someone who doesn’t know the technical vocabulary behind it.
03Built for when things failIf the AI is slow or doesn’t respond, the app stays useful and explains what happened.
03 / how I build
I bring the idea and the judgement. My agents write the code.
I work with several agents and combine them: each one brings something different, and thanks to a shared knowledge base any of them can pick up where another left off. I decide what gets built, review every result and approve what gets published.
agent 01Claude CodeAnthropic
agent 02CodexOpenAI
agent 03AntigravityGoogle
agent 04UltronHermes · local
agent 05VisionHermes · server
directorMatiasIdea · review · approval
04 / about me
From the clinical lab to applied AI.
I started out in clinical laboratory work, where a result is only worth something if the process behind it is reliable: protocols, controls, documentation. Today I apply that same discipline to building tools with AI agents.
I am currently studying a Master’s in Artificial Intelligence at BIG School. I am interested in applied AI systems, agent workflows and automation that demands rigour.
FocusApplied AI · agents · automation
BackgroundClinical lab · QA/QC · SOP
LocationAdelaide now · Europe next
DirectionEurope · Switzerland · Italy