Aditya Oke

I’m a Master’s student in Computational Data Science at Carnegie Mellon University, where I work with Prof. Lei Li in the Li Lab. My research focuses on LLM safety, robustness, and interpretability for agentic and decision-making systems.

Currently, I’m investigating hallucination, attribution, and robustness of large language models like GPT-4, Gemini, and DeepSeek under distribution shift. I’m developing automatic and human-aligned evaluation frameworks for factuality, reasoning consistency, and failure analysis in scientific text generation.

I’m also participating in the Amazon Nova Challenge, where I’m building red-teaming LLM agents to detect vulnerable code generation and creating safety benchmarks for adversarial coding prompts.

Prior to CMU, I worked as a Software Engineer II at JPMorgan Chase, where I designed and deployed distributed Spark pipelines processing multi-terabyte financial datasets and built AWS-based infrastructure for AI-driven compliance tools.

I’m passionate about making AI systems more reliable, interpretable, and safe for real-world applications. I’ve contributed to open-source projects like PyTorch Torchvision and have been recognized with multiple awards for technical excellence and innovation.