Rohit Gupta
Founding AI Engineer at Overmind
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolRohit Gupta is a Founding AI Engineer based in London with eight years of experience building production-grade ML systems, from foundational SLMs to scalable text-to-image services. He has driven backend and platform work—deploying stable-diffusion at scale, designing dynamic batching and microservices to jump concurrency from tens to hundreds, and leading ML platform efforts at PremAI and Mazaal. A prolific open-source contributor and top-5 maintainer on Lightning AI, his enhancements to pytorch-lightning and torchmetrics reflect deep expertise in training optimization, distributed tooling, and reliable model serving. Rohit blends hands-on engineering (FastAPI, MongoDB, Redis, LangChain) with product leadership from multiple founding roles, often owning both product and infrastructure. He is passionate about computer vision and NLP, active in the Python OSS community, and known for turning research techniques into robust developer-facing integrations. Less obvious: he repeatedly scales prototype research into resilient cloud deployments that survive real launch traffic spikes.
8 years of coding experience
5 years of employment as a software developer
Bachelor of Technology - BTech Computer Science, Bachelor of Technology - BTech Computer Science at Maharaja Agrasen Institute Of Technology, Delhi
English, Hindi
Github Skills (17)
pytorch10
python10
machine-learning10
pytorch-lightning10
deeplearning-ai10
deep-learning10
trainings10
model-optimization10
dataloader10
modeling10
metric10
data-science9
unit-testing9
gradient9
test-automation9
Programming languages (9)
Github contributions (5)
Lightning-AI/pytorch-lightning
Apr 2020 - Nov 2022
Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.
Role in this project:
ML Engineer & Back-end Developer
ML Engineer & Back-end DeveloperContributions:5 releases, 3617 reviews, 384 commits in 2 years 7 months
Contributions summary:Rohit's contributions primarily revolve around enhancing the functionality and performance of the `pytorch-lightning` repository, focusing on adding support for new features. These include the addition of test dataloaders, improvements to learning rate logging, and enhancements to model configuration. Further contributions indicate the user’s involvement in optimizing the training process through modifications to gradient accumulation and the integration of automatic mixed-precision training. The user also focused on refactoring existing codebase, including the implementation of the batch size finder, and adding unit tests to ensure correctness.
pythonheadachespytorch-modelsdata-sciencehandling
Lightning-AI/torchmetrics
Jul 2020 - Feb 2022
Machine learning metrics for distributed, scalable PyTorch applications.
Role in this project:ML Engineer
ML EngineerContributions:43 reviews, 11 commits, 3 PRs in 1 year 7 months
Contributions summary:Rohit contributed to the `torchmetrics` repository, focusing on improving metrics related to machine learning. Their work involved fixing parameters, improving documentation, and refactoring code within the `classification` and `regression` modules. They also added new functional regression metrics. The user collaborated with other developers, indicating involvement in code review and integration.
pytorchscalablepythonanalysesdata-science
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.