Junpeng Lao
Staff Data Scientist
Zurich, Zurich, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolJunpeng Lao is a Staff Data Scientist based in Zurich with 10 years of experience bridging cognitive science and production-scale probabilistic modeling. After a PhD in Cognitive Psychology and a postdoc focused on statistical modeling, he joined Google where he rose from Data Scientist to Staff, applying rigorous Bayesian methods to real-world problems. He is an active open-source contributor to flagship probabilistic projects such as TensorFlow Probability, PyMC, BlackJAX and the Bayesian Modeling and Computation in Python book code, improving MCMC diagnostics and NUTS implementations. His background in “brain stuff” gives him a unique edge in designing interpretable models and diagnostics that prioritize scientific validity as well as engineering robustness.
10 years of coding experience
11 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Cognitive Psychology, Doctor of Philosophy (Ph.D.) Cognitive Psychology at University of Glasgow
Stackoverflow
424reputation
5kreached
6answers
1question
Github Skills (30)
bayesian-statistics10
probability10
probabilistic-programming10
python10
bayesian-data-analysis10
testing10
statistics10
machine-learning10
ml10
statistical-inference10
arviz10
mcmc10
pymc10
bayesian10
tensorflow10
Programming languages (9)
Github contributions (5)
BayesianModelingandComputationInPython/BookCode_Edition1
Aug 2021 - Jul 2022
Role in this project:
Data Scientist
Data ScientistContributions:3 reviews, 49 commits, 31 PRs in 10 months
Contributions summary:Junpeng contributed significantly to the code, adding and finalizing chapter 4, 6, 10 and 11 which focus on probabilistic modeling and inference within a Bayesian framework. The user added a variety of code changes. These included changes to import libraries, include model specifications, and examples related to model testing. The contributions suggest an involvement in the practical application of the model examples.
blackjax-devs/blackjax
May 2021 - Jan 2023
BlackJAX is a Bayesian Inference library designed for ease of use, speed and modularity.
Role in this project:Back-end Developer
Back-end DeveloperContributions:9 releases, 389 reviews, 37 commits in 1 year 8 months
Contributions summary:Junpeng contributed to bug fixes and enhancements within the BlackJAX library, focusing on the No-U-Turn Sampler (NUTS) implementation. Their commits addressed issues in trajectory calculations, including U-turn checks and proposal accumulation, and added diagnostic features to monitor the leapfrog counts. They also refactored the progressive integration process to improve efficiency.
speedmodularitybayesian-inferencesampling-methodsprobabilistic-programming
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.