Vittorio Caggiano
Lecturer at Harvard Medical School
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolVittorio Caggiano is a multidisciplinary technology leader and researcher who translates neuroscience insights into practical AI and embodied-intelligence products, currently serving as CTO and Co-Founder of MyoLab while lecturing at Harvard Medical School, King’s College London, and Spaulding Rehabilitation Hospital. With a PhD in neuroscience and nine years of experience across academia and industry, he has first-author publications in Nature, Science, Cell and top ML venues like NeurIPS and ICML, and has led large-scale AI programs at Meta and IBM. He spearheaded infrastructure and tooling efforts used broadly by the community (contributions to FairScale and xFormers) and advanced biomechanical RL environments (MyoSuite), bridging research code to production workflows with integrations like Stable Baselines3 and Weights & Biases. Known for building and directing interdisciplinary teams of researchers, engineers, and designers, he combines deep scientific rigor with hands-on MLOps and product delivery. Notably, his work consistently focuses on motor control and embodied behavior, turning complex neurocomputational models into scalable machine-learning systems and real-world demos.
9 years of coding experience
9 years of employment as a software developer
Master of Science (MS), Electrical and Electronics Engineering, Master of Science (MS), Electrical and Electronics Engineering at Università degli Studi di SalernoPhD, Neuroscience, PhD, Neuroscience at University of Tübingen
Italian, English
Github Skills (18)
pytorch10
distributed-training10
python10
machine-learning10
pipelining10
reinforcement-learning10
pipe10
pipeline10
devops9
mujoco9
documentation9
wandb8
sac8
ppp8
hydra8
Programming languages (4)
Github contributions (5)
MyoHub/myosuite
Feb 2022 - Jan 2023
MyoSuite is a collection of environments/tasks to be solved by musculoskeletal models simulated with the MuJoCo physics engine and wrapped in the OpenAI gym API.
Role in this project:
ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:9 releases, 126 reviews, 114 commits in 10 months
Contributions summary:Vittorio's contributions primarily involve integrating Stable Baselines3 (SB3) for reinforcement learning tasks within the MyoSuite environment. They added job scripts and a launcher script using Hydra for training various RL algorithms like PPO and SAC. Further contributions include integrating WandB for experiment tracking and logging, along with implementing callbacks for evaluation and checkpointing.
physics-engineapiphysicsmusculoskeletal-modelsmusculoskeletal
facebookresearch/fairscale
Sep 2020 - Mar 2022
PyTorch extensions for high performance and large scale training.
Role in this project:MLOps Engineer
MLOps EngineerContributions:37 reviews, 13 commits, 20 PRs in 1 year 5 months
Contributions summary:Vittorio primarily contributed to examples and tutorials related to the `fairscale` library, specifically focusing on demonstrating the use of OSS (Optimizer State Sharding) and pipe (pipeline parallelism) within a PyTorch environment. Their contributions include updating existing tutorial documentation, adding new examples using MNIST, and integrating OSS and pipeline parallelism. They also made modifications to testing scripts, and addressed formatting and import issues, thereby improving code quality and usability within the example implementations.
pytorchdeep-learningmachine-learningscaletraining
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.