Devesh Walawalkar
Research Engineer at UltronAI
Los Angeles, California, United States
Summary
👤
Senior🎓
Top SchoolDevesh Walawalkar is a research engineer specializing in computer vision and deep learning with eight years of experience taking AI systems from research to production across startups and large enterprises. He has published at top venues (ECCV 2024) and built scalable R&D and labeling pipelines that powered commercial products in film dubbing and retail AI, including on-device foundational models for edge deployment. His background in electrical engineering and signal processing informs practical solutions for resource-constrained inference, demonstrated by deploying perception models on Jetson, Qualcomm, and Apple platforms. At Flawless he led self-supervised face identification research and automated face-parsing pipelines using SAM2.1 and Radiov3; at Honeywell and CMU he focused on robotics perception and mobile-ready face-safety applications. Devesh combines hands-on systems engineering with formal research, mentoring interns and winning internal hackathons that bridged prototype research to shipped features. Based in Los Angeles, he maintains an active portfolio (www.devwalkar.info) and a broad interest in deep learning, image processing, and applied AI.
8 years of coding experience
8 years of employment as a software developer
Secondary school, Secondary school at IES Modern English SchoolBachelor of Technology - BTech Electrical and Electronics Engineering, Bachelor of Technology - BTech Electrical and Electronics Engineering at Veermata Jijabai Technological Institute (VJTI)Master's degree Electrical and Computer Engineering, Master's degree Electrical and Computer Engineering at Carnegie Mellon UniversityHigh school certificate Computer Science, High school certificate Computer Science at D.G. Ruparel college for Science, Commerce and Arts
English, French, Marathi, Hindi
Github Skills (6)
lda7
data-storage7
compression6
pruning6
model-compression6
pytorch4
Programming languages (1)
Github contributions (5)
Devwalkar/BOC-KD
Jan 2020 - Aug 2020
Code Base for ECCV Paper: Blockwise Online Model Compression using Knowledge Distillation
Contributions:35 commits, 3 pushes, 1 branch in 7 months
pytorchdistillationknowledge-distillationldaknowledge
Devwalkar/website
Nov 2020 - Jul 2023
Contributions:17 pushes, 1 branch in 2 years 8 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.