Registered user since Wed 27 Mar 2019
Senior Applied Scientist at Microsoft MAIDAP Team. PhD in Computer Science (W&M, US), MS in Computer Engineering (TUM, Germany & UNAL, Colombia), BS in Computer Engineering (UNAL, Colombia). David was part of the SEMERU Research Group under Professor Denys Poshyvanyk. His research focused on interpretable deep learning for code generation, with a particular emphasis on using causal inference to explain software generative models. Professionally, he’s worked on projects involving refactoring automation, traceability link recovery using IR and DL, and security classification with pre-trained models. During internships at Cisco and Microsoft, David explored vector representations for software artifacts and developed interpretability methods for neural code models. He also brings several years of back-end software engineering experience.
Contributions
2026
ICSME
ICSE
- Author of Mapping the Trust Terrain: LLMs in Software Engineering — Insights and Perspectives within the Journal-first Papers-track
- Author of A Causal Perspective on Measuring, Explaining and Mitigating Smells in LLM-Generated Code within the Research Track-track
- Author of Enabling Global, Human-Centered Explanations for LLMs: From Tokens to Interpretable Code and Test Generation within the Research Track-track