Chapter I
About
Shipping production software since 2020, from clinical research platforms to consumer apps. I care most about the parts that break quietly.
Since 2021 I have been a senior full-stack engineer at ViewMind, where I lead three production web apps for clinicians and the research data platform underneath them. Alongside it I build ski and outdoor-sports products with a small founding team: SnowMatch, bebetter and Sparta. Before that I co-founded One Percent and was its CTO from 2023.
A master's in Big Data Management and Analytics at UPC Barcelona, where I took crypto price forecasting from on-chain data and public text to BERT sentiment, LSTM and XGBoost models, and a dashboard. Now I build tools for AI agents that fail in production.
A wrong number in a clinical report is a production incident.
Chapter II
From data to system
Four stages, one person who owns all of them.
01 · Data
Get the schema right.
Research data that still holds together years into a longitudinal study. Hourly on-chain metrics joined with Reddit, news and tweets. Quality and query speed first.
ViewMind · CryptoVoice
02 · Model
Measure, then trust.
Batched transformer inference, bidirectional LSTMs and XGBoost compared across market regimes, scored with MAE and MSE on held-out data. Detectors that label why an agent failed.
CryptoVoice · Forensi
03 · System
Ship it to people.
Native SwiftUI and Ionic clients, NestJS and FastAPI backends, serverless on AWS and Cloudflare. Something a clinician, a skier or a crew can actually use.
One Percent · Sparta · bebetter · Brick and Brew
04 · Operate
Own what breaks.
Review, CI and releases for a small team. Billing consistency, audited agent tools, and production failures handled by the person who built the system.
ViewMind ops MCP · One Percent · SnowMatch
Chapter III
The work
Seven products, current role first. Results first, story second.
- 01 ViewMind Leads three production web apps for clinicians and research staff, and the research data platform underneath them. 2021–now
- 02 SnowMatch Find, compare and book certified ski and snowboard instructors, with the tools ski schools need behind it.1,000+ instructors · 20+ centres 2022–now
- 03 Sparta The social layer for outdoor sports: post a session, fill it with friends, show up together.2.4k users · 41% D30 retention 2026
- 04 bebetter AI ski coaching: upload a run and get technique corrections, with optional review by elite instructors.160+ riders · 150+ video analyses 2025–2026
- 05 Forensi Root-cause analysis for production AI agents: the likely cause, the evidence behind it, and the fix. 2026
- 06 One Percent A subscription wellness app I co-founded and built the engineering for, as CTO (2023–2026). 2023–2026
- 07 Brick and Brew A private scoreboard for a crew who run, ride and swim hard, then have beers together. 2026
Chapter IV
Experience
The short version.
- 2021–now
Senior Full-Stack Engineer · ViewMind Inc.
Leads three production web apps for clinicians and research staff, and designed the research data platform underneath them.
- 2022–now
Ski & outdoor products · SnowMatch, bebetter, Sparta
Ski and outdoor-sports products built with Tomás Bacigalupo and Ignacio Fernández Capiet. Frontend Lead on SnowMatch, co-founder of bebetter (AI ski coaching) and Sparta (CPO). React, Java, Expo, SwiftUI, FastAPI on AWS.
- 2023–2026
Co-founder & CTO · One Percent
Co-founded subscription wellness app. I ran the engineering: Angular/Capacitor client, NestJS API, RevenueCat billing on AWS, plus review, CI and releases.
- 2023–2024
MSc Big Data Management & Analytics · UPC Barcelona
Thesis: CryptoVoice. Price forecasts for eight crypto assets from hourly market structure and public text.
- 2021
Software Engineer · Province of Buenos Aires
Internal systems under government security and compliance constraints.
- 2020–2021
Full-Stack Developer · Glotools
Customer SaaS in React, Ruby on Rails and SQL, released with GitHub Actions.
- 2016–2021
Information Systems Engineering · UTN Buenos Aires
Degree.
Production
- TypeScript
- React
- Angular
- NestJS
- SwiftUI
- Ruby on Rails
- AWS
- Docker
- GitHub Actions
Models & data
- Python
- pandas
- TensorFlow/Keras
- XGBoost
- Hugging Face
- MongoDB
- SQL
Chapter V
Notes
Field notes from what I ship. Checklists, not theory.