LatentWorlds AI

Over LatentWorlds AI

LatentWorlds AI is building the deployment backend for physical AI.

Our product is a robotics-native data and operations platform that captures multi-modal recordings, retrieves synchronized sensor and log slices, and produces reproducible datasets for debugging and training.

Robotics teams lose too much time wrangling data pipelines instead of building their core product. We fix that, starting with reliable edge-to-cloud capture over unreliable networks and extending into incident replay, dataset versioning, and fleet-scale operational workflows. We deploy in managed cloud, customer VPC, hybrid, or on-prem environments, and integrate into your existing stack rather than replacing it.

What we do:

- Multi-modal capture and synchronized retrieval across sensors, state, and logs - Incident-to-replay workflows that give engineering teams exact context, fast - Versioned, traceable dataset exports with full provenance for reproducible training - Auditable operations with RBAC, retention controls, and audit trails

LatentWorlds AI was founded by Cristian Meo, CEO, who completed his PhD cum laude at TU Delft in Generative AI with a focus on world modeling and model-based reinforcement learning, and Alejandro Daniel Noel, CTO, who spent 3.5 years at Google Cloud in Zurich where he built a product now deployed in millions of cars worldwide.

We are currently onboarding design partners for 2026 pilots. If your team operates robot fleets in the field and needs reliable data infrastructure, get in touch!

Adres

• Hoofdkantoor

  Molengraafsingel 122629JD DelftZuid-Holland, Nederland

Team per functiegebied

Gebaseerd op 1 openbare functieprofielen, stand van 3 september 2026.

• Other / Unclassifiable1