Feed
Pro
Search

Author

Write
Drafts

Bug0 - The AI-native e2e QA regression testing Passmark - The open-source AI framework for regression testing Hackathons Changelog Brand Hashnode gql skill - let your AI agent publish to your Hashnode blog @hashnode on X Hashnode on LinkedIn Support - hello+support@hashnode.com Code of Conduct Terms Privacy Sitemap
Sign in

Search Hashnode

Search posts, tags, users, and pages

@tanguy-pauwels

Tanguy Pauwels

@tanguy-pauwels·France·Joined April 2026

ML Engineer & Full Stack Developer

About

From precision mechanics at BMW to high-performance Machine Learning. Focused on optimization, raw performance, and architectural pragmatism.

Available for

Collaborating on latent space analysis, robotics data engineering, and scaling JEPA architectures.

Tanguy Pauwels's blogs

Technical Blueprintsblog.tanguy-pauwels.fr3 posts

About

From precision mechanics at BMW to high-performance Machine Learning. Focused on optimization, raw performance, and architectural pragmatism.

Available for

Collaborating on latent space analysis, robotics data engineering, and scaling JEPA architectures.

Tanguy Pauwels's blogs

Technical Blueprintsblog.tanguy-pauwels.fr3 posts

Articles Comments

Recently published

TPTanguy Pauwelsinblog.tanguy-pauwels.fr·Apr 15 · 11 min read

[3] Post-Mortem Analysis: Why My First World Model (JEPA) Is "Blind"

The pipeline is ready, the data is converted, and the GPU has completed its first "stress test." Now comes the most critical phase for any research engineer: the autopsy of the latent space. In this t

TPTanguy Pauwelsinblog.tanguy-pauwels.fr·Apr 15 · 4 min read

[2] Data Engineering: Why and How to Convert LeRobot (Parquet/MP4) to HDF5

In my previous post, I explained why the JEPA architecture is such a promising lead for robotics. But between Yann LeCun’s theory and the first \(loss.backward()\), there is a massive wall: the data.

TPTanguy Pauwelsinblog.tanguy-pauwels.fr·Apr 15 · 5 min read

[1] Rethinking Robotics: Why I’m Betting on JEPA over VLAs

To test my research intuitions, I’m documenting my work on a JEPA (Joint-Embedding Predictive Architecture) world model, starting with low-cost robotics datasets like LeRobot and the Koch SO-ARM101. T