Applied Scientist · Research × Engineering
I study hard problems and build useful things.
My work joins AI research with engineering, from the first experiment to reliable systems people can use in the real world.
Recommendation Systems · Temporal Graph Learning · ML Systems
Treverse Recsys: Temporal Recommendations from Learning to Serving
An end-to-end recommendation system combining inductive temporal graph retrieval, reproducible model releases, in-memory decision serving, and statistically controlled experiments.
Take a look ↗Selected work
I follow research questions all the way to working systems.
Multimodal AI · Vision-Language Models
Inverse-LLaVA: Rethinking Multimodal Alignment via Text-to-Vision Mapping
Inverse-LLaVA projects text embeddings into continuous visual space and fuses the modalities within intermediate transformer layers, enabling multimodal reasoning without explicit alignment pre-training.
Take a look ↗
Natural Language Processing
AI for Negotiation
LLM systems that code negotiation transcripts and simulate realistic negotiation scenarios for research.
Take a look ↗
Computer Vision
Ancient Mortars Classification
A Vision Transformer pipeline for classifying ancient mortar particles and comparing material composition across archaeological samples.
Take a look ↗Beyond the work
A little more about me.
See how I got here, what keeps me curious, or meet Mino, my coding companion.
Visitor map