Akshit Khurana
Engineering at Ramp
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAkshit Khurana is a seasoned engineering leader with 14 years of experience building scalable AI and data platforms, currently leading engineering at Ramp in San Francisco. He co-founded and served as CTO of Narrative (YC W23), where he designed an LLM-driven data automation platform and secured enterprise customers and VC backing. Previously he led PyTorch Mobile work at Facebook, contributing NNAPI integration and performance wins that materially improved on-device model throughput, and remains an active contributor to the core PyTorch repo. His background spans full-stack and platform engineering—from streaming pipelines and microservices at Ambient.ai to large-scale ML data pipelines and computer vision at HouseCanary—combining hands-on coding with hiring and team scaling. Notably, he’s maintained side projects like a static site generator and has a history of shipping both open-source contributions and production-grade systems that bridge research and product engineering.
14 years of coding experience
12 years of employment as a software developer
Indian Institute of Technology Roorkee
Stackoverflow
684reputation
828kreached
3answers
4questions
Github Skills (28)
git10
pytorch10
python10
machine-learning10
deep-learning10
github-pages10
tensor10
aws-dynamodb9
wget9
autograd9
dynamodb9
amazon-dynamodb9
vulkan8
html8
gpu8
Programming languages (9)
Github contributions (5)
axitkhurana/buster
Oct 2013 - Oct 2014
Brute force static site generator for Ghost
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:26 commits, 1 comment, 2 issues in 1 year
Contributions summary:Akshit primarily developed a static site generator for Ghost blogs. They built the core functionality, allowing users to generate, preview, and deploy static pages to Github Pages. The user also implemented features like adding custom domains, and packaging the project as a Python module. Furthermore, they integrated the use of PyQuery to fix HTML and link issues, and improved the software with bug fixes and code style improvements.
static-siteghostsite-generatorbrute-forcebrute
pytorch/pytorch
Sep 2020 - Nov 2022
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:ML Engineer
ML EngineerContributions:41 reviews, 87 commits, 62 PRs in 2 years 1 month
Contributions summary:Akshit primarily contributed to the PyTorch framework, focusing on enhancing its capabilities for model compilation and optimization. Their work included adding new APIs for NNAPI integration, specifically addressing compilation preferences and float16 relaxation. Furthermore, the user made critical fixes related to tracing and profiling within the Dynamo module and corrected how Tensor.T is handled. They also addressed a Vulkan context initialization issue and added configuration updates for the torch.ao.nn module.
pythongpu-accelerationdeep-learninggpunumpy
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.