BRKGA for Multi-Depot Orienteering Problem
Implemented a Biased Random Key Genetic Algorithm for an NP-complete optimization problem, outperforming VNS-based baselines.
Backend & AI Engineer
A selection of projects I have worked on over the years.
Implemented a Biased Random Key Genetic Algorithm for an NP-complete optimization problem, outperforming VNS-based baselines.
Developed a RAG-based code review agent using AST-based chunking and semantic analysis to generate fast, high-quality inline pull request feedback.
A Tetris game whose AI weights are evolved live by a Biased Random-Key Genetic Algorithm (BRKGA). Built with a pure NumPy engine and a heuristic AI doing 2-ply search; the population keeps evolving whenever the AI tops out.