gsoc idea list and contribution guide

[midaa1]

Hello everyone,
I am a machine Learning Engineer with solid experience in classical machine learning and deep learning, including CNNs and RNNs, using PyTorch. Skilled in building, training, and optimizing models for real-world applications, with a strong understanding of data preprocessing, feature engineering, and model evaluation.

I was asking about the idea list and the contribution rules

[Hellboy28D]

Dear @zeydabadi ,

Greetings of the day.

I am Divakar Daya, a 2nd-year Data science student at the Indian Institute of Technology, Madras. I am writing to express my interest in contributing to Emory University’s Department of Biomedical Informatics as a GSoC 2025 contributor.

I love working on applied Machine Learning and AI systems, particularly in building end-to-end, production-oriented ML pipelines. I have hands-on experience designing multi-agent AI systems, Retrieval-Augmented Generation (RAG) workflows, and data-driven predictive models, and I enjoy working at the intersection of research ideas and deployable software.

Recently, I have worked on:

• A satellite-powered air quality forecasting platform using atmospheric and satellite data to predict pollution trends
• Advanced multi-agent AI systems using LangGraph, RAG, and Langflow, with modular agents and graph-based orchestration
• AI-driven web scraping and information extraction pipelines using Python, LLMs, and Streamlit

My technical stack includes Python, ML/AI, LLM systems, data analysis, and backend development, with experience using tools such as Docker, Git/GitHub, Streamlit, and modern ML frameworks. I am comfortable spending time deeply understanding existing codebases, research context, and design decisions before contributing meaningful pull requests.

I have attached my CV and included relevant links below for reference:

• GitHub: https://github.com/Hellboy28D
• LinkedIn: https://www.linkedin.com/in/divakar-daya-55178034b/
• Kaggle: https://kaggle.com/competitions/agents-intensive-capstone-project/writeups/new-writeup-1764149505136
• Mail: yadavdayaram.nuh@gmail.com

I would be excited to contribute steadily to Emory BMI’s open-source efforts and align my work closely with the group’s long-term goals. If there are particular repositories or components where you believe sustained contributions would be most valuable, I would be happy to start there.

Thank you for your time and for supporting contributors in the open-source community.

Best regards,
Divakar Daya
IIT Madras

[TiyaaaJain]

Hey everyone,

I'm Tiya, I am drawn to areas around deep learning, AI/ML deployment, and I would love to learn and contribute to this organisation. I’m an undergraduate student in my third year of college, and I’m really looking forward to learning, building, and growing this summer through open-source collaboration.