Charchit Sharma
Senior Machine Learning Engineer at IDfy
Mumbai, Maharashtra, India
Summary
👤
Senior🎓
Top SchoolCharchit Sharma is a Senior Machine Learning Engineer based in Mumbai with nine years of experience bridging research and production ML systems. He has worked on large-scale 3D and 2D-to-3D research at Avataar.ai and as a research assistant at IIT Hyderabad under Prof. Vineeth N. Balasubramanian, and now applies that expertise at Idfy. His open-source contributions include performance and compatibility improvements to Hugging Face's widely used diffusers library, notably adding torch.compile support and cross-platform fixes for advanced diffusion training pipelines. Comfortable moving models from experimental code to robust training scripts, he combines academic rigor with practical engineering from his time at Infosys and earlier industry projects. A maker at heart, he maintains an active GitHub and personal site documenting his work and experiments.
9 years of coding experience
5 years of employment as a software developer
High School, Science, High School, Science at Army Public School 39 GTC, VaranasiBachelor of Technology - BTech, Computer Science, Bachelor of Technology - BTech, Computer Science at Swami Keshwanand Inst. Of Tech. Mgt. & Gramothan,Jaipur
English, Hindi
Github Skills (8)
net10
pytorch10
machine-learning10
stable-diffusion10
deep-learning10
python10
flax4
jax3
Programming languages (7)
Github contributions (5)
huggingface/diffusers
Sep 2023 - Dec 2024
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.
Role in this project:
ML Engineer
ML EngineerContributions:11 reviews, 21 PRs, 95 comments in 1 year 3 months
Contributions summary:Charchit made significant contributions to the diffusers library, primarily focused on integrating and improving training scripts and pipelines related to diffusion models. They introduced and adapted features for torch.compile compatibility across multiple training scripts including ControlNet, InstructPix2Pix, and T2I Adapter for both base and SDXL models, demonstrating a focus on optimization and performance. The user also refactored core components such as image_processor.py and made adjustments for broader platform compatibility including Windows support.
pytorchartdeep-learningimage2imagestate-of-the-art
charchit7/charchit7
May 2021 - Jan 2025
Contributions:35 pushes, 1 branch in 3 years 8 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.