Tim Sainburg
Cambridge, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolTim Sainburg is a research scientist and machine learning engineer with 11 years of multidisciplinary experience at the intersection of biosignals, computational neuroscience, and software engineering. Currently a postdoc at Harvard and working with Meta, he builds end-to-end neurobehavioral pipelines—spanning hardware, electrophysiology, and large-scale 3D rodent kinematics—and has led an 18-species mouse kinematic dataset project. He developed and productionized ML tools such as a parametric UMAP variant and the widely used noisereduce library, contributing both model innovations (global structure losses, sampling strategies) and robust engineering (CI/CD, GPU support, large-dataset handling). Tim combines rigorous experimental design (award-winning PhD work in cognitive science) with practical full-stack development skills, from TensorFlow generative models to deployment-ready testing and documentation. Notably, his background ranges from fieldwork with primates to building biologically realistic auditory models, giving him a rare blend of hands-on neuroscience and scalable ML engineering.
11 years of coding experience
8 years of employment as a software developer
University of California, San DiegoBS, Psychology - Neuroscience, BS, Psychology - Neuroscience at Penn State University
Github Skills (22)
notebook10
dimensionality-reduction10
python10
imap10
testing10
machine-learning10
cicd10
tensorflow10
omap10
generative-model10
jupyter-notebook10
autoencoder9
keras9
librosa8
numpy8
Programming languages (9)
Github contributions (5)
timsainb/noisereduce
Mar 2019 - Sep 2022
Noise reduction in python using spectral gating (speech, bioacoustics, audio, time-domain signals)
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:7 releases, 6 reviews, 42 commits in 3 years 7 months
Contributions summary:Tim primarily contributed to the setup and improvement of the project's infrastructure, including CI/CD integration with Travis CI. They also implemented several functional testing, added and updated the documentation, and configured the testing environment, including adding GPU acceleration support and updating the versioning and testing process to reduce noise. The user demonstrates significant contributions to the project's maintainability and development workflow.
audio-processingpythonperlin-noiseaudioreduction
timsainb/tensorflow2-generative-models
May 2019 - Nov 2020
Implementations of a number of generative models in Tensorflow 2. GAN, VAE, Seq2Seq, VAEGAN, GAIA, Spectrogram Inversion. Everything is self contained in a jupyter notebook for easy export to colab.
Role in this project:ML Engineer
ML EngineerContributions:31 commits, 7 pushes, 5 comments in 1 year 6 months
Contributions summary:Tim primarily contributed to implementing and updating various generative models within the TensorFlow 2 framework. Their work includes adding autoencoders and iterators related to the NSYNTH dataset and addressing JSON-related errors in the notebooks. Furthermore, the user improved the project's documentation and user experience by updating colab links and metadata in the notebooks. The user also implemented and refined the WGAN-GP model and adjusted colab dependencies for the GAIA model, demonstrating a focus on model training and deployment.
inversioneverythingself-containedspectrogramjupyter-notebook
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.