Nima Sedaghat
Deep Learning Chair Active Asteroids NASA Partner Program at University of Washington
Seattle, Washington, United States
Summary
👤
Senior🎓
Top SchoolNima Sedaghat is a deep learning and computer vision researcher with 11 years of experience leading AI efforts at the intersection of astronomy and industry. Based at the University of Washington, he directs teams building the computer vision pipeline that will detect supernovae in images from the LSST/Rubin telescope and leads the Active Asteroids NASA partner program combining deep learning with citizen science. His work spans from convolutional encoder-decoders to generative models and vision-language systems, recently tuning DeepSeek’s Janus-Pro to interpret astronomical imagery. Former roles at ESO and OSRAM saw him pioneer generative models across 1D–3D astronomical data and apply ML to industrial sensing, reflecting a rare blend of scientific curiosity and production-grade engineering. He holds a PhD in Computer Vision from the University of Freiburg and is known for getting generative models to learn astrophysical features without explicit supervision.
11 years of coding experience
12 years of employment as a software developer
Doctor of Philosophy (PhD), Computer Vision, Doctor of Philosophy (PhD), Computer Vision at The University of FreiburgAmirkabir University of Technology
English, German, Persian
Github Skills (71)
caffe10
loss-functions9
deep-learning9
pytorch9
machine-learning9
optical-flow9
twitter-api9
twitter8
notebook8
astronomy8
keras8
jupyter-notebook8
orbital-mechanics7
astrology7
science7
Programming languages (7)
Github contributions (5)
lsst-dm/dmtn-216
Jan 2022 - Dec 2022
Deep Learning Approach(es) for LSST Alert Production
Contributions:4 reviews, 2 PRs, 51 pushes in 10 months
pytorchalertlsstdeep-learningapproach
lsst/meas_transiNet
May 2022 - Feb 2024
Contributions:106 reviews, 24 PRs, 182 pushes in 1 year 10 months
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.