About

The data-ops team for people actually building embodied AI.

Train Them AI recruits, trains and manages a global network of camera operators who record first-person video — silent or narrated step by step — of real household, commercial and industrial work, and delivers it QC'd and training-ready to robotics and AI-data teams that can't afford Scale AI's $500K minimums but need the same quality.

Why we exist

The bottleneck for most embodied-AI teams isn't model architecture. It's data: diverse, real-world, first-person footage of people doing real tasks in real environments — collected to a protocol, QC'd every day, and shipped in the exact format the training loop expects. Large vendors lock that work behind enterprise contracts. Academic labs can't move at startup speed. And building your own collection network means running recruiting, training, QC and payroll across a dozen countries — an ops company inside your ML team.

We sit in the middle: an operations-first, format-native team that already runs that company — Train Them AI LLC, registered in Wyoming, USA, with operations led from Córdoba, Argentina. 1,000+ active operators in 20+ countries, 30+ team leads, a dashboard that handles onboarding, uploads, QC feedback and twice-monthly payouts, and a QC pipeline that combines automated checks — including speech-to-text and LLM scoring on narrated clips — with human review on every batch. First QC'd pilot batch in two weeks, in LeRobot, RLDS, HDF5 or your format.

Who's behind it

Diego Pousa, founder of Train Them AI
Diego Pousa
Founder & CEO · Córdoba, Argentina
Rodrigo Dominguez, co-founder and head of operations at Train Them AI
Rodrigo Dominguez
Co-Founder · Head of Operations
Alan Pousa, operations manager at Train Them AI
Alan Pousa
Operations Manager · Recruiting, Pods & QC

What we've shipped

1,000+
Active paid operators
25,000+
Hours delivered
2 wk
Brief to delivery
20+
Countries, 4 continents

How we work

No hardware minimums. No 6-month sales cycles. Pilots start at 100 hours. We slot into your pipeline wherever you need us — collection, narration, annotation, or all of it — on your capture app or ours. Samples first, then a protocol you sign off on, then rolling weekly deliveries in your target format with a full data card and QC report. Your feedback on batch N shapes batch N+1.

Start a Pilot → Become a Collaborator