Who am I?

André Silva

PhD Student in Computer Science at KTH Royal Institute of Technology

I work mostly on machine learning for program repair. My PhD topic is differentiable program repair: we translate programs into neural networks to perform repair at the numerical level.

Other than that, I have done work on fine-tuning LLMs for program repair both in supervised and self-supervised settings, and have developed benchmarks that are representative of real-world tasks and can be used to evaluate the latest frontier models.

andreans@kth.se GitHub LinkedIn X (Twitter) KTH Profile Google Scholar