An Exploratory Study of Bayesian Prompt Optimization for Test-Driven Code Generation with Large Language Models
Uses Bayesian optimization to improve test-driven LLM code generation across HumanEval+ and multiple code models.
Ethan Villalovoz
Sacramento, California
I'm an M.S. student in Computer Science at Georgia Tech. I'm interested in how robots can understand people and the physical world, make decisions under uncertainty, and adapt through interaction.
Previously, I built and evaluated enterprise AI agent systems at Microsoft.
You can reach me at .
Publications
Social Triangles and Aggressive Lines: Multi-Robot Formations Impact Navigation and Approach
Shows that triangular robot formations are perceived more positively and navigate more robustly than line formations.
Experience
Microsoft
Software Engineer Intern
May 2026 – Jul 2026
CMU Robotics Institute
Summer Scholar
Jun 2024 – Aug 2024
STEP Intern
May 2023 – Aug 2023
Oregon State University
NSF REU Fellow
Jun 2022 – Aug 2022
Teaching
Georgia Institute of Technology
Graduate Assistant
Fall 2026
Washington State University
Undergraduate Teaching Assistant
Spring 2025
CPT S 315: Introduction to Data Mining
Undergraduate Teaching Assistant
Fall 2024
CPT S 350: Design and Analysis of Algorithms
Undergraduate Teaching Assistant
Fall 2023
CPT S 355: Programming Language Design
Undergraduate Teaching Assistant
Fall 2022
CPT S 121: Program Design and Development in C