I am an AI researcher and engineer with a proven track record of translating cutting-edge research into real-world, high-impact AI products. I specialize in multimodal AI, large-scale embeddings and retrieval systems, and agentic AI, delivering innovations from prototype to production. Recently, I’ve been building multimodal AI systems at Meta that combine vision-language models, agentic capabilities, and efficient inference to create video-based always-on, seamless user experiences on wearable devices.

Previously, at eBay, I led the computer vision team and rebuilt the entire visual search engine—from large-scale training data collection, golden dataset curation and annotation, distributed training and inference system creation, to backfill pipelines and real-time indexing, enabling world-class visual search, recommendations, and magical listing systems at scale. My work directly impacted product discovery, personalization, and inventory management.

Over the past decade, I’ve led projects in image, video, and text indexing, foundation models, object recognition, segmentation, image generation, and AI-driven products, always bridging research and production to deliver tangible impact.

Outside of AI, I enjoy painting as a creative outlet and soccer, which fuels my passion for teamwork and collaboration—skills I bring to cross-functional engineering teams every day