Hi, I'm
Freddy Boulton
I'm a software engineer who loves building tools at the intersection of open source and ML. Right now I'm at Hugging Face, where I am the Team Lead and a core maintainer of Gradio and FastRTC — open-source libraries that make it easier to build and share ML applications.
Experience
Places I've worked and things I've built along the way.
Hugging Face
2022 – PresentSoftware Engineer
- › Core maintainer of Gradio, an open-source Python library for building ML web apps (35k+ GitHub stars, 1.1M+ monthly users)
- › Created FastRTC, a real-time communication framework for Python AI apps
- › Built the Gradio custom components ecosystem, enabling third-party extensions
- › Developed real-time streaming features for audio/video ML applications
Alteryx
2020 – 2022Software Engineer
- › Core developer on EvalML, an open-source AutoML library
- › Implemented automated ML pipeline search and optimization
- › Built time series forecasting capabilities
Motional (nuTonomy)
2018 – 2020Data Scientist
- › Developed prediction models for autonomous vehicle motion planning
- › Built tools for large-scale driving data analysis
Nielsen
2017 – 2018Data Scientist
- › Applied statistical modeling to media measurement data
- › Built data pipelines for audience analytics
What I Build
Open-source tools I've created or help maintain.
Gradio Custom Components
Designed and built the plugin ecosystem for Gradio, enabling developers to create and share custom UI components for ML apps.
EcosystemEvalML
Open-source AutoML library for building, optimizing, and evaluating machine learning pipelines with domain-specific objective functions.
Open SourceTalks & Writing
I like sharing what I learn — at conferences, in blog posts, and in tutorials.
Conference Talks
Unlock the Future of AI: Real-Time Voice & Video
Building real-time AI applications with FastRTC and WebRTC
MLOps World 2025Gradio: The Web Framework for Humans and Machines
How Gradio bridges the gap between ML models and end users
ODSC East 2024Building Machine Learning Demos with Gradio
A practical tutorial on creating and sharing ML applications
Mastering LLMs 2024Mastering LLMs Conference
Guest speaker on building interactive ML demos
Writing & Publications
Gradio MCP Servers
Turning Gradio apps into Model Context Protocol servers
HF BlogFastRTC + Cloudflare
Global real-time AI infrastructure with Cloudflare's TURN network
HF BlogGradio Reload Mode
Instant UI iteration during Gradio development
Real PythonMemory Management in Python
A case study on debugging memory problems with open source tools
PaperCoverNet: Multimodal Behavior Prediction using Trajectory Sets
arXiv · 2019
PaperMotion Prediction using Trajectory Sets and Self-Driving Domain Knowledge
arXiv · 2020
Education
University of Chicago
M.S. in Statistics · 2017
B.A. in Statistics · 2016
Thesis: Statistical methods for large-scale data analysis
By the Numbers
From the Blog
Things I've been thinking about and learning lately.