Yogesh Kumar
Founding ML Engineer at Solid
Berlin, Germany
Summary
👤
Senior🎓
Top SchoolYogesh Kumar is a Founding ML Engineer with a decade of experience building production-grade machine learning systems, currently focused on video intelligence from his base in Berlin. He blends deep research pedigree—a PhD in Computer Science from Aalto and prior studies at NYU and UCSD—with hands-on engineering, having built Nokia’s proprietary PyTorch+MLflow engine and multi-GPU training pipelines. He has led CV and agentic LLM efforts at startups, shipping an automated image-enhancement pipeline that couples diffusion models with autonomous prompt-generation. An active open-source contributor, he improved reliability and logging in the popular pytorch/ignite library and added Visdom integration and targeted test coverage. His research output includes sample-efficient training for ViTs and LLMs, interpretability work on CKA metrics, and an attention-free model for EHR prediction, showing a rare mix of systems, models, and applied ML in healthcare and edge domains. Colleagues describe him as an engineer who literally "multiplies large matrices on a GPU for a living"—a succinct hint at his low-level performance focus alongside product impact.
10 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Aalto UniversityMaster's degree Computer Science, Master's degree Computer Science at New York UniversityUniversity of California, San DiegoBachelor's degree Mechanical Engineering, Bachelor's degree Mechanical Engineering at Savitribai Phule Pune University
Github Skills (11)
neural-network10
pytorch10
machine-learning10
python10
metric10
testing10
deeplearning-ai9
deep-learning9
documentation8
cicd7
tensorboard7
Programming languages (7)
Github contributions (5)
pytorch/ignite
Feb 2020 - Mar 2020
High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.
Role in this project:
ML Engineer & Test Automation Engineer
ML Engineer & Test Automation EngineerContributions:7 commits, 11 PRs, 23 comments in 1 month
Contributions summary:Yogesh primarily contributed to the improvement and maintenance of the `pytorch/ignite` library. Their work includes bug fixes, such as synchronizing the progress bar with epoch counts and correcting issues in the tqdm logger. They added and updated tests, demonstrating a focus on ensuring the library's reliability and functionality. Furthermore, the user integrated a Visdom logger and improved documentation within the repository.
pytorchpythondeep-learninghigh-levelneural-networks
ykumards/ykumards
Jul 2020 - Feb 2025
Contributions:13 pushes, 1 branch in 4 years 7 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.