VIP Program: AlphaEvolve Subteam

Georgia Institute of Technology · Automated Algorithm Design · Atlanta, GA · Jan 2026 - Present

Research Focus

In the VIP program, I contribute to a GTRI-sponsored research initiative exploring multi-objective genetic programming systems that evolve hybrid algorithms directly from data.

Core Contributions

• Evolutionary Frameworks: Working on systems that combine search, mutation, and objective evaluation to discover high-performing algorithm variants.
• LLM-Guided Evolution: Extending workflows where LLMs propose and mutate code candidates in an automated evolution loop.
• Experiment Rigor: Building reproducible Python experiment pipelines with traceable settings, run metadata, and consistent evaluation criteria.
• HPC Execution: Running and validating experiments on Georgia Tech's PACE ICE infrastructure for neural architecture and performance studies.

Why It Matters

This experience sits at the intersection of AI systems engineering and scientific experimentation, and it has sharpened my skills in optimization, reproducibility, and analysis at scale.