Akhil Mathur
AI Research Scientist, Meta Superintelligence Labs
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolAkhil Mathur is an AI Research Scientist at Meta Superintelligence Labs with seven years of experience advancing LLM reasoning, agents, memory, and reinforcement learning, and was a core contributor to Llama 3. Previously a Principal Research Scientist and Distinguished Member of Technical Staff at Bell Labs, he led teams on federated, distributed, and on-device ML, filing 15 patents and publishing widely. He holds a PhD in Machine Learning from UCL, an MS from the University of Toronto, and earned the President’s Gold Medal at DA-IICT. Akhil contributes to prominent open-source federated learning tooling—enhancing the Flower framework with QffedAvg and benchmark expansions—and has had research covered by outlets like the New Yorker and Financial Times. His profile blends deep academic rigor with pragmatic systems work, particularly on efficient deployment and domain adaptation for resource-constrained and real-world settings.
7 years of coding experience
12 years of employment as a software developer
MS Computer Science, MS Computer Science at University of TorontoBachelor of Technology - BTech Computer Science, Bachelor of Technology - BTech Computer Science at Dhirubhai Ambani UniversityUniversity College London
Hindi, English
Stackoverflow
13reputation
1kreached
0answers
2questions
Github Skills (15)
pytorch10
machine-learning10
tensorflow10
deep-learning10
python10
federated-learning10
ai9
artificial-intelligence9
benchmark8
benchmarking8
imagemagick6
nodejs6
proxy6
numpy6
image-processing6
Programming languages (1)
Github contributions (5)
adap/flower
May 2020 - Jan 2022
Flower: A Friendly Federated AI Framework
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 18 commits, 8 PRs in 1 year 8 months
Contributions summary:Akhil contributed significant enhancements to the `flower` framework, a federated learning framework. They implemented a QffedAvg baseline strategy, demonstrating a focus on advanced federated learning techniques. The user also modified settings related to federated learning benchmarks, showcasing an understanding of practical experiment configurations. Furthermore, the user expanded the benchmark suite to include a TensorFlow-based spoken keyword classification model.
federated-analyticsfederated-learning-frameworkmachine-learningkeras-federated-learningflower
akhilmathurs/libriadapt
Aug 2020 - Aug 2021
Contributions:33 commits, 32 pushes, 1 branch in 11 months
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.