Hello! 👋

I'm Sasi Kiran Gaddipati — a Doctoral Researcher at TIB Hannover, working on AI for science and LLM reasoning. My research focuses on building systems that support scientific discovery: from human-AI collaborative research workflows to benchmarking autonomous AI systems on machine learning reproducibility.

Previously, I worked as a Data Scientist at ensun GmbH, applying NLP and machine learning to industry-scale problems.

On this site you'll find my research publications, technical write-ups, and thoughts on AI, LLMs, and the future of scientific discovery. When I'm not working, I enjoy watching chess and films, and following new ideas at the edge of AI and science.

• MLReplicate: Benchmarking Autonomous Research Systems for Machine Learning Reproducibility

  Published: 15 May, 2026 |  at  02:00 AM

  MLReplicate is an end-to-end benchmark evaluating autonomous research systems on machine learning reproducibility, constructed from ICML 2025 outstanding papers and assessing 6 state-of-the-art AI research systems.

  research AI LLM Benchmarking

• AISSISTANT: Human-AI Collaborative Review and Perspective Research Workflows in Data Science

  Published: 8 Mar, 2026 |  at  01:00 AM

  AIssistant is the first open-source agentic framework for Human-AI collaborative generation of scientific perspectives and review research in data science. Accepted to PAKDD 2026 - Hong Kong, China.

  research AI LLM Research Workflows

• The Real-World Curriculum

  A practical education resource covering skills rarely taught in school — from emergency response and personal finance to digital safety, mental health, and career literacy. Structured from beginner to advanced.

  Education Life Skills Open Source