AI Research Scientist | PhD @ Dartmouth
I build and evaluate multimodal LLMs that actually understand video and vision. My research focuses on step-verified reasoning, cross-modal fusion, and long-video understanding—plus designing benchmarks and datasets that push these models forward.
👨💻 What I work on:
- Multimodal LLMs (vision-language) & computer vision
- Video understanding, moment retrieval & evaluation frameworks
- Large-scale AI datasets (specs → tooling → QA → release)
- Production ML systems with distributed training/inference
🛠️ Tech stack: PyTorch • JAX • TensorFlow • CUDA • Kubernetes • Docker • AWS/GCP • SQL/NoSQL • C++ • Python
🌎 Background: Colombian computer scientist now based in the US. I've shipped large-scale geo-spatial systems, worked with teams across LATAM and the US on data analytics and ML pipelines, and contributed to Python open-source projects.