Kunal Banerjee
Principal AI Scientist at Walmart Global Tech
Bengaluru, Karnataka, India
Summary
👤
Senior🎓
Top SchoolKunal Banerjee is a Principal AI Scientist with 9+ years of experience translating cutting-edge research into production-grade AI systems across Walmart and Intel Labs. He specializes in Responsible GenAI—monitoring LLMs for accuracy, bias, hallucination and policy drift—while also delivering solutions in CV, NLP, recommendation, AutoML and AI governance. His work spans end-to-end impact: prototyping PoCs, presenting to leadership, and scaling successful ideas into planet-scale offerings used across Walmart and its subsidiaries. At Intel he contributed high-performance convolution implementations (Winograd) to prominent open-source libraries like LIBXSMM and MKL-DNN, enabling faster deep learning on Intel architectures. He holds a PhD from IIT Kharagpur with award-winning thesis work and has published and demonstrated low-precision networks (ternary residuals) and BFLOAT16 efficacy in top venues. Colleagues rely on him for a rare Research-to-Realization mindset that pairs deep systems know-how with practical, measurable business outcomes.
9 years of coding experience
8 years of employment as a software developer
Higher Secondary, Higher Secondary at Ramakrishna Mission Residential CollegeICSE, ICSE at LyceeDoctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at Indian Institute of Technology, KharagpurBachelor of Technology (B.Tech.) Computer Science and Engineering, Bachelor of Technology (B.Tech.) Computer Science and Engineering at Heritage Institute of Technology, Kolkata
Github Skills (6)
c1710
avx10
convolution10
assembly10
c1110
mathematical-optimization9
Programming languages (3)
Github contributions (5)
libxsmm/libxsmm
Feb 2017 - Aug 2019
Library for specialized dense and sparse matrix operations, and deep learning primitives.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:310 commits, 81 PRs, 8 comments in 2 years 6 months
Contributions summary:Kunal appears to be implementing new features for convolution using Winograd. This is evident from the addition of support for KNM and fixing various compilation issues related to the Winograd functionality. The user's contributions include changes to core files related to assembly code generation for the AVX512 architecture, particularly for Winograd convolution, and also involved optimization efforts.
bfloat16avxsimdlapackavx512
kunalbanerjee/kunalbanerjee.github.io
Feb 2017 - Jan 2025
Contributions:257 pushes, 1 branch in 8 years 1 month
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.