Ankit Mathur
Software Engineer at Databricks
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAnkit Mathur is a software engineer with six years of experience building production ML and model-serving infrastructure, currently leading Model Serving for GPUs and LLMs at Databricks in San Francisco. He holds EECS from UC Berkeley and a CS master's from Stanford, and brings research-grade rigor from DAWN and RISE lab projects into product-focused engineering. His open-source contributions to MLflow improved Pyfunc model saving/loading and GPU support for PyTorch and Keras, directly enhancing model serving usability at scale. Ankit also scouts for Greylock, blending technical product insight with early-stage investing in ML, data, and cloud startups. He’s comfortable spanning research, systems design, and hands-on coding—often surfacing subtle interoperability and performance fixes that make complex ML stacks reliably deployable.
6 years of coding experience
3 years of employment as a software developer
Bachelor of Science (BS), Electrical Engineering and Computer Science, Bachelor of Science (BS), Electrical Engineering and Computer Science at University of California, BerkeleyMaster's degree, Computer Science, Master's degree, Computer Science at Stanford University
English, Hindi, Spanish
Github Skills (7)
continuous-deployment10
keras10
pytorch10
machine-learning10
mlflow10
python10
ml-deployment10
Programming languages (7)
Github contributions (5)
mlflow/mlflow
Jun 2019 - Jan 2022
Open source platform for the machine learning lifecycle
Role in this project:
ML Engineer
ML EngineerContributions:99 reviews, 36 commits, 88 PRs in 2 years 6 months
Contributions summary:Ankit primarily focused on enhancing the MLflow library's Pyfunc functionality. Their commits addressed critical aspects of model saving, loading, and logging for PyTorch and Keras models, including adding support for custom objects and GPU utilization. The contributions included modifications to existing methods such as `log_model` and `save_model`, and involved writing tests to ensure new features worked correctly. The user's work improved the usability and performance of MLflow's model serving capabilities, particularly for models leveraging GPU resources.
pythonlifecyclemlmachine-learningincremental-learning
ankit-db/mlflow
May 2020 - Jan 2022
Open source platform for the machine learning lifecycle
Contributions:4 PRs, 144 pushes, 33 branches in 1 year 7 months
machine-learning-lifecyclepythonlifecyclemachine-learningmlops
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.