Dev Khant is a founder and machine learning engineer with 6 years of experience building production-first ML systems and developer tools, now serving as Co-founder & CTO of Steadwing—an autonomous on-call engineer that diagnoses and remediates production incidents in minutes. Previously the Founding ML Engineer at Mem0 (YC S24), he led backend and infra for a high-throughput service and maintained an open-source repo with tens of thousands of stars and millions of downloads. His work spans MLOps, CV, NLP and recent applied research on LLMs, including a co-authored paper accepted to ECAI 2025 and a 400+ citation footprint. A Kaggle 4x Expert and active open-source contributor, he’s implemented multi-backend support (JAX, Paddle) for the ivy project and has integrations across major developer tools. Dev combines customer-facing product work with low-latency systems engineering—experience that shows in serving 8k req/s backends and onboarding tens of thousands of engineers to new platforms. Based in the Bay Area, he’s passionate about turning ambitious research ideas into practical, production-ready systems.
5 years of coding experience
4 years of employment as a software developer
Higher Secondary School, Higher Secondary School at Modi School
Bachelor of Technology - BTech, Computer Science, Bachelor of Technology - BTech, Computer Science at Darshan University
Stackoverflow
Stats
1reputation
0reached
0answers
0questions
Github Skills (13)
ivy10
paddle10
pytorch10
machine-learning10
jax10
tensorflow10
python10
numpy9
deeplearning-ai9
deep-learning9
converter8
transpiler7
transcode7
Programming languages (8)
TypeScriptSCSSJavaScriptVueHTMLJupyter NotebookRich Text FormatPython
Contributions:11 PRs, 24 comments, 12 issues in 4 months
Contributions summary:Dev contributed to the implementation of JAX and Paddle backend support for the Ivy library. The contributions include the addition of new functions, such as `isscalar`, `csingle`, and `cdouble` within the JAX backend, demonstrating work on supporting a new frontend. Moreover, the user resolved conflicts and updated the codebase related to Paddle backend, highlighting their contributions to overall framework compatibility.
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.