Kevin Chen
Founder at Tempora Labs
Washington, District of Columbia, United States
Summary
🤩
Rockstar🎓
Top SchoolKevin Chen is a founder and data scientist with a decade of experience building ML-driven products and developer tools, currently leading Tempora Labs from Washington, D.C. He blends hands-on engineering—from contributing uncertainty-aware conformal prediction enhancements to the popular NeuralProphet forecasting library—to applied analytics and product strategy at Fannie Mae. His background spans blockchain and agent frameworks at Fetch.ai and IOTA, where he shipped proofs-of-concept for delivery tracking and developer tooling, reflecting a knack for bridging distributed systems with practical business use cases. Trained in computer science (University of Virginia) with specialized nanodegrees in deep RL and ML engineering, he pairs rigorous technical depth with a founder’s appetite for prototyping and adoption. An often-overlooked thread in his career is his sustained focus on developer-facing integrations—APIs, agent connectors, and example repos—that accelerate real-world usage of complex technologies.
10 years of coding experience
8 years of employment as a software developer
Economics Commerce, Economics Commerce at University of OxfordNanodegree Deep Reinforcement Learning, Nanodegree Deep Reinforcement Learning at UdacityBachelors in Computer Science, Bachelors in Computer Science at University of Virginia
English, Chinese
Github Skills (10)
neural-network10
forecasting10
pytorch10
machine-learning10
forecast10
time-series10
uncertainty10
deep-learning10
estimate10
python10
Programming languages (6)
Github contributions (5)
ourownstory/neural_prophet
Aug 2022 - Jan 2023
NeuralProphet: A simple forecasting package
Role in this project:
ML Engineer
ML EngineerContributions:2 releases, 109 reviews, 73 commits in 5 months
Contributions summary:Kevin's primary focus was on enhancing the uncertainty estimation capabilities of the `neural_prophet` forecasting package. They implemented and refined conformal prediction methods, including "naive" and "CQR" techniques, to generate prediction intervals. Their contributions involved modifying the core `forecaster.py` file, the example notebook, and related testing files, demonstrating a deep understanding of machine learning and time-series forecasting. These changes directly improved the model's ability to quantify and communicate forecast uncertainty.
forecastingneuralprophetpythonforecasttime-series
Photrek/Coupled-VAE
Aug 2020 - Aug 2021
Machine learning algorithms for complex systems leveraging the methods of nonlinear statistical coupling
Contributions:47 commits in 1 year
complex-systemsleveragingnonlinearcouplingmachine-learning-algorithms
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.