Gowthami Somepalli — Co-Founder
Hi there! I am a multimodal researcher based out of SF Bay area. My research focuses on multimodal understanding (Vision LLMs) and generative modeling (diffusion-based image generation), with recent work on mid- and post-training optimization of text-to-image and image editing models using RLHF, DPO, and GRPO. During my PhD, I studied and mitigated memorization in generative models, resulting in publications at NeurIPS, CVPR, and ECCV. My research is well recieved in research community (> 2500 citations, h-index 15) and in applied ML community (~ 1k Github stars across repos). Checkout my google scholar for full list of publications - https://scholar.google.com/citations?user=T2ezBDsAAAAJ&hl=en
Stackforce AI infers this person is a generative modeling expert in the SaaS industry.
Location: Mountain View, California, United States
Experience: 9 yrs 4 mos
Skills
- Generative Modeling
- Generative Ai
- Machine Learning
- Product Management
Career Highlights
- Published research at NeurIPS, CVPR, and ECCV.
- Over 2500 citations and h-index of 15.
- Expertise in multimodal understanding and generative modeling.
Work Experience
World Labs
Member of Technical Staff (2 mos)
Adobe
Research Scientist (1 yr 1 mo)
Meta
Research Scientist Intern (7 mos)
Amazon Web Services (AWS)
Applied Research Scientist Intern (2 mos)
University of Maryland
Graduate Research Assistant (5 yrs 4 mos)
Flipkart
Manager - Business Development (8 mos)
Poolka
Co-Founder (2 yrs 1 mo)
Education
Doctor of Philosophy - PhD at University of Maryland
Bachelor of Technology (B.Tech.) at Indian Institute of Technology, Madras
Master of Technology (M.Tech.) at Indian Institute of Technology, Madras
