Welcome, I'm Saaketh Raghava

A passionate programmer and researcher interested in Mechanistic Interpretability, LLM Evaluation, and creative technologies.

Timeline

2025
Present Day
November 2024
Research paper on LLM hallucinations Published!
October 2024
Released Coffi
May 2024
Started as Undergraduate Researcher at UC Merced, working on LLM Evaluation
June 2023 - August 2023
SAP Intern at Asahei Kasei North America, streamlined client order management processes
March 2023 - June 2023
Ambassador at Inspirit AI, developed AI algorithm to reduce racial bias in criminal profiler databases

Projects

Screenshot of Coffi application

Coffi

California Bills summarization application
Visualization of Lumina

Lumina

Mini LLM Implementation in C++

Research

Classification of Hallucinations in Large Language Models Using a Novel Weighted Metric

Supported by UC CITRIS and the Banatao Research Institute

Read Paper

ATC Pilot QA Dataset

Description

Contact

Things I find cool!

Misc.

My favorite readings and recommendations

Catastrophic AI Risks State of AI 2024