Divam Gupta
Founder at Stellon Labs Inc
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolDivam Gupta is a founder and machine learning researcher with 11 years of experience building applied ML systems across industry and academia, from Microsoft Research to Meta (AI for VR). He holds an MS in Robotics from Carnegie Mellon and has first-authored papers at AAAI and ICLR, reflecting a strong research-to-product trajectory. At Meta he focused on social presence in VR, and earlier roles covered unsupervised learning, disentangled representations, and stereo-to-BEV perception for robotics. Divam is an active open-source contributor—his TensorFlow/Keras implementation of Stable Diffusion and the one-click Diffusion Bee UI demonstrate fluency across ML model internals and end-user tooling. He combines deep technical rigor with product instincts, often translating research ideas into widely usable libraries and consumer-friendly interfaces. Based in San Francisco, he blends startup grit with research pedigree to tackle real-world AI problems.
10 years of coding experience
4 years of employment as a software developer
Master of Science - MS Robotics, Master of Science - MS Robotics at Carnegie Mellon University
Stackoverflow
228reputation
17kreached
3answers
2questions
Github Skills (26)
javascript10
python10
electron10
image-processing10
machine-learning10
user-interface10
stable-diffusion10
keras10
deep-learning10
tensorflow10
vue10
vue-jsx10
css9
html9
nlp8
Programming languages (11)
Github contributions (5)
divamgupta/diffusionbee-stable-diffusion-ui
Sep 2022 - Jan 2023
Diffusion Bee is the easiest way to run Stable Diffusion locally on your M1 Mac. Comes with a one-click installer. No dependencies or technical knowledge needed.
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:23 releases, 3 reviews, 198 commits in 4 months
Contributions summary:Divam's contributions primarily focused on implementing the user interface (UI) for the Diffusion Bee application. The commits added UI elements and features, including the creation of a gallery item component and other user interface elements. The user also integrated functionalities for managing the application, such as the setting of settings, and the integration of image saving and sharing functionalities.
dependenciesstablediffusiondiffusionimage-generationbee
divamgupta/stable-diffusion-tensorflow
Sep 2022 - Nov 2022
Stable Diffusion in TensorFlow / Keras
Role in this project:ML Engineer
ML EngineerContributions:44 commits, 15 PRs, 36 pushes in 2 months
Contributions summary:Divam primarily contributed to the implementation of the Stable Diffusion model in TensorFlow/Keras. Their work focused on defining and building various components of the model, including CLIP text embeddings, transformer layers, and the UNet model. They also integrated a tokenizer for processing text inputs and added functionality for image-to-image generation. Their commits demonstrate a deep understanding of the model architecture and its underlying components.
deep-learningkeras-tensorflowdiffusiontensorflowkeras
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.