Live
Tailored CVs per posting, tracked applications, and the eligibility check up front instead of after you have already applied.
Buenos Aires, Argentina · open to remote AI product & AI engineering roles
Strategy, product and code.
I think it and I build it.
Agents, automation and whole products from nothing. I think them through, I build them, and I put them in front of people. Weeks, not quarters.
I go after the systems nobody designed badly on purpose. They just got old and nobody touched them.
- Products built
- 14
- Years in software
- 10+
- Years on my own products
- 4
- Tools delivered
- 2,500+
- Contributions, last year
- 4,845
MBA en Tecnología e Innovación, en curso
Work
Products I built and still run. All of them are live, and you can walk into every one.
Live
An AI tutor for kids in LatAm. Nova builds missions around what a kid already loves, and never hands over the answer.
Live
Daily carpooling in Buenos Aires. A flat prepaid fare that does not spike at rush hour, matched to drivers already making that trip.
Live
Where early projects get found. Builders publish what they are working on and meet co-founders before there is anything to demo.
Live
Specs, roadmaps, research and audits for PMs, founders and small product teams. Product management built on Claude.
Live
AI agents for small businesses: WhatsApp, web and voice, wired to the data they already have, with a dashboard on top.
Also built
Earlier builds. Two are still up, the rest are off, and another four ran their course and came down.
-
01
Live
ComoVenimos
A community for tech and startup people, with a blog and paid memberships.
-
02
Off
Fiteando
A social app for finding a training partner at your gym, at your hour.
-
03
Off
Book Translated
Translates and summarises whole books the same day, priced by word count.
-
04
Off
Valmidor
Daily gamified challenges with rankings and a WhatsApp loop.
How I got here
Before the current title, mostly in rooms where being wrong cost more than a rollback. In order.
-
Consulting
Process consulting for industry
Before software: walking an operation and mapping how the work actually moves through it against how the manual says it does, then finding the step that costs the most. Nothing about that question changed when the subject became a product.
-
Health
Cardiology and clinical systems
QA on systems where a wrong reading is not a bug report. It is where the habit started: I only trust a system I can test, and I would rather find the failure than be told it cannot happen.
-
Security
On-premise, locked down, Linux
QA and on-premise deployments for enterprise clients with closed networks, simulating their Linux environments to test against and running the releases myself. Everything that ran twice got scripted.
-
Energy
IoT and hardware in the field
Devices and energy systems that live outside the datacentre, where a release you cannot roll back is a truck roll. Testing against real hardware behaviour instead of a mock that always agrees with you.
-
Logistics
Routing, fleets, and a team
Product for route optimisation and fleet operations. Built the prototyping practice from nothing and led it, short enough loops that a working thing showed up in the conversation instead of a slide.
-
AI product
Agents from zero to production
Architecture for agent systems and the roadmap over them: what the agent may do, where it escalates, how it is evaluated. Led the team that built the workflow builder those agents run on, and shipped features into production alongside it.
Craft
What I am genuinely good at.
AI agent products
Agent systems end to end, across chat, WhatsApp and voice. The hard part is never the model, it is what the agent is allowed to do.
Agent architecture
The system the agents live in, zero to production, and the team that builds it: tool boundaries, escalation, what counts as a good answer.
Agentic coding
Claude Code and Cursor every day. I write production features and ship them, so I do not just prioritize work, I move it.
Voice and multimodal
Voice LLMs on live calls, and images as input rather than decoration: reading a screenshot or a photo and deciding from it.
Generative production
Image generation and product video, prompt to rendered clip, on a pipeline. A landing gets its visuals the week it gets its copy.
Automation and internal tooling
n8n, Activepieces, Retool, Zapier, Make. The boring machinery that hands a team its hours back, and the process that keeps it from rotting.
Growth and instrumentation
Umami, LogRocket, session review, Meta Ads. Counting the funnel from the database and not the dashboard, because the dashboard lies.
0 to 1
Standing up a function, a team or a product where none existed, then designing how it runs without me in the middle. More than once.
Quality and release engineering
Six years in QA and release automation before product: test design, on-premise deployments, pipelines. It is why I only trust systems I can test.
Data and BI
SQL, Tableau, Metabase. Dashboards people actually make decisions on, not screenshots for a monthly deck.
Stack
- Python
- SQL
- Claude API
- Claude Code
- Cursor
- n8n
- Activepieces
- Retool
- Zapier
- Make
- Supabase
- Vercel
- Coolify
- Tableau
- Metabase
- Jenkins
- Umami
- LogRocket
- Hedera
Work with me
The same things, pointed at someone else's problem. Companies, startups, and people with a project that has not started yet.
Strategy and advisory
You have the engineers. What is missing is someone deciding what the AI should not do, and saying no to the demo that will not survive a user. One to one with a founder, or alongside the team.
If you have engineers and the AI decisions are being made by whoever is free.
Agent systems, spec to production
What the agent is allowed to do, where it hands off to a person, and how you find out when it stops being good. Chat, WhatsApp and voice, and the ones that read an image and decide from it.
If somebody on your team spends the day moving the same data between two screens.
0 to 1
An idea to something people can use. Spec, build, release, in weeks. You end up owning the repo.
If you have the idea and no team to build it.
Your business online, and something that answers
A landing built or redesigned so it says what you do, with the measurement wired in from the start, and a chat agent trained on your business so somebody who arrives at three in the morning gets an answer.
If your site has not changed since you made it and you do not know how many people arrive.
Automation and internal tooling
n8n, Activepieces, Retool, Zapier, Make. Plus the process that keeps it from rotting after I leave, which is the part that goes missing.
If the process lives in a spreadsheet that only one person understands.
Tell me what is broken and I will tell you whether I am the right person for it. If I am not, I will say so, which is cheaper for both of us than finding out in week three.
Start a conversationAgent Skills
The skills I write for Claude Code. Six a product or engineering team would reach for, and the rest underneath.
ship
The full loop for a feature: plan, implement, validate, review, deploy. Chains the other skills together.
feature-agents
Same as ship, delegated to five subagents with narrow tool access: the one writing the spec cannot write code.
arquitecturasprivate
Picks the agent architecture a task deserves out of 32 methods, and returns the one to use, why not the others, and the signal to abandon it.
review
Code review of a diff. Every finding anchored to a real file and line, with the concrete failure scenario spelled out.
test
Runs your validation pipeline and reports honestly: what passed, what failed, what got skipped. Never marks green something that never ran.
deploy
The release pipeline. Separate permissions for commit, push and deploy, and it never ships a build that hasn't passed.
The other thirteen, dashed ones are not published
- project-setup
- dev
- deep-review
- ux
- product
- deploy-qa
- decision-log
- design-system
- community-manager
- web-cinematica
- video-producto
- voz-auditar
- one per product
Activity
Most of the code lives in private product repos, so the graph is the honest part of it.
What is public is this site and the Agent Skills above. Everything else is product code in private repos, which is where most of that graph comes from. See the profile on GitHub
Let's talk
Open to remote AI roles, open to building for you, and always up for a conversation about agents, automation, or why the commute is still broken. Leave it here and it reaches my phone in a few seconds.
