Albert Bou
Research Scientist in AI at FutureHouse
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAlbert Bou is a Deep Reinforcement Learning researcher and applied ML engineer with eight years of experience bridging academic rigour and industry impact, now based in San Francisco. He holds a PhD from Universitat Pompeu Fabra where he developed RL solutions for real-world problems and contributed to Meta’s TorchRL—fixing core optimizer and PPO issues and adding new objectives—demonstrating strong open-source and production-oriented chops. His work spans drug discovery with LLM-driven RL and current research on reasoning agents that plan and make decisions in complex environments. Colleagues describe him as objective and results-driven, combining a researcher’s curiosity with practical engineering to push ML from prototypes into deployable systems.
8 years of coding experience
4 years of employment as a software developer
Bachelor's degree (Exchange Program) Telecommunications Engineering, Bachelor's degree (Exchange Program) Telecommunications Engineering at Polytechnique MontréalMaster's programme in Machine Learning Mechine Learning, Master's programme in Machine Learning Mechine Learning at KTH Royal Institute of TechnologyThe University of MelbourneUPC Universitat Politècnica de CatalunyaDoctor of Philosophy - PhD Deep Reinforcement Learning, Doctor of Philosophy - PhD Deep Reinforcement Learning at Universitat Pompeu Fabra
English, French, Catalan, Spanish
Github Skills (6)
pytorch10
machine-learning10
ppp10
python10
reinforcement-learning10
documentation7
Programming languages (2)
Github contributions (5)
pytorch/rl
Nov 2022 - Jan 2023
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
Role in this project:
ML Engineer
ML EngineerContributions:79 reviews, 14 commits, 84 PRs in 2 months
Contributions summary:Albert primarily contributed to bug fixes and feature enhancements within the PyTorch Reinforcement Learning library. Their work involved addressing issues in core components such as optim steps and PPO objectives, modifying code related to model definition (ConvNet), and adding new functionalities like an A2C objective class. The user also added documentation to helper functions and classes, making the library more accessible.
pytorchpythonreinforcement-learningprimitivedeep-reinforcement-learning
PyTorchRL/rl
Nov 2022 - Nov 2024
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
Contributions:1 PR, 987 pushes, 130 branches in 2 years
pytorchpythonreinforcement-learningprimitivedeep-reinforcement-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.