Ayush Thakur
Machine Learning Engineer at Weights & Biases
Kolkata, West Bengal, India
Summary
🤩
Rockstar🎓
Top SchoolAyush Thakur is a machine learning engineer based in Kolkata with eight years of professional experience and over five years focused specifically on building ML systems. Currently at Weights & Biases, he contributes to the core wandb platform—integrating telemetry for libraries like CatBoost, TensorFlow/Keras, and LightGBM, and adding metrics logging and checkpointing that help move experiments toward production. As a former CTO and co-founder of an open-source community, he blends product-minded engineering with community-driven development and reproducible research practices. His background in electronics and communication gives him a hardware-aware perspective on model deployment and performance tuning. Known for updating example projects to showcase practical ML workflows, he helps lower the barrier for practitioners to adopt robust experiment tracking.
8 years of coding experience
2 years of employment as a software developer
Bachelor of Technology, Electronics and Communication Engineering, Bachelor of Technology, Electronics and Communication Engineering at Netaji Subhas Engineering College
English
Stackoverflow
468reputation
16kreached
12answers
0questions
Github Skills (28)
data-visualizations10
visualization10
python10
scikit10
machine-learning10
data-visualisation10
tracks10
keras10
deeplearning-ai10
deep-learning10
tensorflow10
scikit-learn10
experiment10
data-visualization10
visualizations10
Programming languages (9)
Github contributions (5)
wandb/examples
Sep 2020 - Jan 2023
Example deep learning projects that use wandb's features.
Role in this project:
ML Engineer
ML EngineerContributions:8 reviews, 25 commits, 20 PRs in 2 years 4 months
Contributions summary:Ayush primarily focused on integrating and utilizing the Weights & Biases (wandb) library within various scikit-learn machine learning projects. They updated existing example scripts across different scikit-learn modules such as regression, classification, and clustering to include wandb logging for metrics, model visualization, and experiment tracking. Furthermore, they added and modified wandb callbacks for logging and model saving within a LightGBM project and a Keras project, showcasing integration with advanced training and evaluation tools.
pytorchdeep-learningdeep-learning-examplemachine-learningwandb
wandb/wandb
Apr 2021 - Jan 2023
The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.
Role in this project:ML Engineer
ML EngineerContributions:69 reviews, 30 commits, 43 PRs in 1 year 9 months
Contributions summary:Ayush primarily contributed to the integration and enhancement of the Weights & Biases platform, specifically focusing on the integration of new machine learning libraries. Their work involved adding telemetry for various libraries like CatBoost, and TensorFlow/Keras, as well as adding a metrics logger and model checkpointing callbacks. This included modifications to the `wandb/proto` files for new features and adjusting code to align with newer versions of TensorFlow and LightGBM.
pythoncollaborationtensorflowhyperparameter-tuningcli
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.