Summary
Eric-tuan Lê is a Member of Technical Staff at Anthropic with eight years of research-driven experience in deep learning, computer vision, and 3D/4D generative modeling. Previously an AI Research Scientist at Meta, he led multimodal Llama post-training work—combining structured-image tuning, large-scale synthetic QA, and SFT→RL pipelines—to improve stepwise reasoning and consistency on text+image tasks. His PhD from UCL and internships at Adobe and Snap produced ICCV/CVPR papers and patents on primitive fitting, 3D mesh recovery, and temporal 4D consistency, and he has strong hands-on expertise in flow-matching over SDFs and score-distillation methods. Comfortable at the intersection of research and engineering, he scales dataset generation and post-training pipelines as readily as he designs geometry-aware losses for real-time systems. He reviews for top venues (CVPR/ICCV/ECCV/NeurIPS), reflecting active engagement with the research community and rigorous experimental evaluation.
8 years of coding experience
4 years of employment as a software developer
Master of Research in Computer Vision & Machine Learning (Master MVA), Computer Vision and Machine Learning, Master of Research in Computer Vision & Machine Learning (Master MVA), Computer Vision and Machine Learning at ENS Paris-Saclay
Master of Science in Applied Mathematics, Master of Science in Applied Mathematics at CentraleSupélec
University College London
Exchange Program, Exchange Program at Singapore Management University
Preparatory classes for French Grandes Ecoles, GPA: 4.00, Mathematics and Physics, Preparatory classes for French Grandes Ecoles, GPA: 4.00, Mathematics and Physics at Lycée Hoche
Master In Management, Master In Management at ESCP Business School
French, English, Spanish, Chinese