Ravin Kohli
Machine Learning Engineer at CUJU
Dusseldorf, North Rhine-Westphalia, Germany
Summary
🤩
Rockstar🎓
Top SchoolRavin Kohli is a Machine Learning Engineer with a decade of experience bridging research and production to deliver practical deep learning systems, currently applying computer vision and ETL pipelines to football performance analytics at Cuju.ai in Düsseldorf. He co-developed and maintained Auto-PyTorch during graduate research, contributing search-space and pipeline improvements that strengthened automated hyperparameter tuning for PyTorch models. His background spans large-scale experiment orchestration (SLURM/MOAB), integrating optimizers like Optuna and SMAC3, and memory optimizations to scale training—skills he leverages to move models from prototype to deployment. Early work at CERN and bio-imaging internships shows a knack for turning noisy scientific data into robust ML predictions, and his fluency in both research methodology and production tooling makes him effective at translating domain expertise into measurable metrics for scouts and coaches.
10 years of coding experience
6 years of employment as a software developer
Masters Computer Science (Specialisation: Artificial Intelligence), Masters Computer Science (Specialisation: Artificial Intelligence) at The University of FreiburgMount Carmel School, Anand NiketanBachelor’s Degree Computer Software Engineering, Bachelor’s Degree Computer Software Engineering at Delhi College of Engineering
English, Hindi, German
Stackoverflow
153reputation
8kreached
0answers
20questions
Github Skills (15)
hyperparameter-optimization10
pytorch10
machine-learning10
deep-learning10
datatable10
automl10
python10
tabular10
django-urls6
django-models6
date-formatting6
appstore6
rest6
django6
django-templates6
Programming languages (1)
Github contributions (5)
automl/Auto-PyTorch
Feb 2021 - Jan 2023
Automatic architecture search and hyperparameter optimization for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:3 releases, 918 reviews, 365 commits in 1 year 11 months
Contributions summary:Ravin made significant contributions to the `auto-pytorch` repository, focusing on updating the search space for hyperparameter optimization. Their work involved adding and testing search space updates, including those for hyperparameter ranges, and modifying pipeline components such as the base pipeline, feature preprocessors and loss strategies. They also implemented changes related to embedding layers and addressed issues with weighted loss.
hyperparameter-optimizationpytorchautomldeep-learningtabular-data
ravinkohli/Auto-PyTorch
Jun 2020 - May 2023
Automatic architecture search and hyperparameter optimization for PyTorch
Contributions:4 PRs, 571 pushes, 99 branches in 2 years 11 months
pytorchhyperparameterdeep-learningoptimizationmachine-learning
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.