Ikechukwu “Ike” Uchendu
About
I am a doctoral student at Harvard University in the Edge Computing Lab, co-advised by Vijay Janapa Reddi and Kianté Brantley, and a Kempner Institute Graduate Fellow. Prior to this, I worked as an AI Resident at Google Brain.
Previously, I received a BS/MS in computer science from Michigan State University.
My main research interest is in how we can better use foundation models as agents in the real world that can improve over time with more experience.
Publications
- See it to Place it: Evolving Macro Placements with Vision-Language Models NeurIPS 2026
- QuArch: A Benchmark for Evaluating LLM Reasoning in Computer Architecture ICML 2026
- QuArch: A Question-Answering Dataset for AI Agents in Computer Architecture IEEE Computer Architecture Letters 2025
- A2Perf: Real-World Autonomous Agents Benchmark arXiv 2025
- Jump-Start Reinforcement Learning ICML 2023
- ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design ISCA 2023
- Scalable Diagnostic Screening of Mild Cognitive Impairment Using AI Dialogue Agent Scientific Reports 2020
- Improving Mild Cognitive Impairment Prediction via Reinforcement Learning and Dialogue Simulation arXiv 2018
Teaching
Teaching Assistant, Harvard University
- CS249r: Tiny Machine Learning, Fall 2023
- Contributor, Machine Learning Systems, the open-source textbook from the course
Teaching Assistant, Michigan State University