Sean Narenthiran
Senior Deep Learning Scientist at NVIDIA
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolSean Narenthiran is a Senior Deep Learning Scientist with 11 years of experience building production-ready NLP and speech recognition systems, currently contributing to NVIDIA after leading research engineering at Grid AI and Digital Reasoning. He blends deep research instincts with software engineering pragmatism—shipping scalable ASR pipelines, integrating PyTorch Lightning, and optimizing multi-GPU training for real-world deployments. An active open-source contributor, Sean has improved high-profile projects like PyTorch Lightning, NeMo and deepspeech.pytorch, focusing on robustness, testing, and diarization for speech models. He pairs a First Class Computer Science degree from King’s College London with a background in game engine content creation, a detail that hints at a long-standing passion for systems, tooling and creative engineering.
11 years of coding experience
6 years of employment as a software developer
Bachelor’s Degree, Computer Science, First Class Honours, Bachelor’s Degree, Computer Science, First Class Honours at King's College London
Github Skills (24)
pytorch10
python10
clustering-algorithm10
testing10
machine-learning10
pytorch-lightning10
deep-learning10
trainings10
generative-ai10
asr10
neural-network10
speech-recognition10
modeling10
model-checking9
checkpoint9
Programming languages (9)
Github contributions (5)
SeanNaren/deepspeech.pytorch
Oct 2020 - Apr 2021
Speech Recognition using DeepSpeech2.
Role in this project:
ML Engineer
ML EngineerContributions:7 releases, 9 reviews, 14 commits in 6 months
Contributions summary:Sean primarily contributed to the integration of PyTorch Lightning within the DeepSpeech model training pipeline, which included adding autocast support, fixing multi-GPU support, integrating checkpointing, and updating the training configuration. They updated smoke tests and pretrained models and corrected class usage, indicating a focus on testing and model deployment. Additionally, the user made minor fixes and updates to configuration files for inference and training.
deepspeech2recognitionspeechspeech-recognition
Lightning-AI/pytorch-lightning
Oct 2020 - Dec 2022
Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.
Role in this project:ML Engineer & Test Automation Engineer
ML Engineer & Test Automation EngineerContributions:8 releases, 1247 reviews, 95 commits in 2 years 2 months
Contributions summary:Sean primarily focused on testing and improving the PyTorch Lightning library. They added tests to ensure the correct score is reported in the checkpoint filepaths, expanded testing for torch save serialization with DDP, and fixed a bug related to logged metrics within test_epoch_end. The user also addressed bugs related to accumulation of gradients and max steps.
pythonheadachespytorch-modelsdata-sciencehandling
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.