Dear Sir/Madam,
I hope you’re doing well. My name is Ujjwal Sharma, a third-year student at the Indian Institute of Information Technology, Bhopal. I am inspired by Kubeflow’s mission to streamline ML model deployment and scaling on Kubernetes, and I am very excited about the possibility of contributing as part of Google Summer of Code (GSoC) 2026.
I would love to start contributing even before the official GSoC period to gain practical experience and make meaningful contributions. I am eager to understand the Kubeflow codebase, explore areas where my skills can add value, and learn best practices for deploying ML pipelines at scale.
I have some experience with Python, Go, TypeScript, Kubernetes, and YAML, and I have built projects involving AI, automation, and scalable full-stack systems:
1. LLM Powered Code Accelerator– Built an AI tool to optimize Python into C++, achieving up to 60,000× speedup on compute-heavy tasks, integrating LLMs via HuggingFace + Gradio.
2. Stock Prediction Portal– Full-stack Django + React app with LSTM-based forecasting, serving 5,000+ API requests/month and visualizing 10+ years of data.
3. Automate the Boring Stuff– Developed 6 automation tools with Django + Celery + Redis, handling 1M+ records asynchronously.
4. UrbanKart – Scalable Django platform deployed on AWS with PostgreSQL, S3, and PayPal integration.
Portfolio: https://ujjwal-sharma-portfolio.netlify.app/
I am curious about a few things and would greatly value your guidance:
For someone new to Kubeflow, which areas are best suited for beginners—building ML pipelines, microservice management, or deployment automation?
Should I focus initially on Python-based operators and ML workflows, or explore Go components and Kubernetes integrations?
Are there opportunities to contribute toward generative AI integrations, dynamic scaling, or improving user-centric features?
I am eager to learn, contribute effectively, and help advance Kubeflow’s goal of making scalable ML pipelines accessible and user-friendly.
Thank you for your time, and for the incredible work you are doing to simplify ML deployments across diverse infrastructures.
Warm regards,
Ujjwal Sharma