multimodal-data
(4)
Mikayel Khachatryan Put a Price on Human-Made Data
Wirestock began by removing the paperwork from stock photography. Its CEO is now trying to solve a larger bargain: how creators get paid when their work helps machines learn.
Elliott Green
Elliott Green is the co-founder and CEO of Dandelion Health, a New York-based real-world data and clinical AI platform building multimodal, longitudinal datasets from a consortium of non-academic health systems. A former founding member of Oscar Health with a finance background, Green launched Dandelion in 2020 to give life science companies and AI developers unbiased, clinical-grade data - including the images, waveforms, and notes that make up most of medicine but are rarely used. The company closed a $14M Series A in 2026 and has launched a free AI validation service and a GLP-1 data library aimed at bringing precision medicine to cardiometabolic disease.
Richard Meng
Richard Meng is the co-founder and CEO of Roe AI, a Y Combinator-backed (W24) startup building an AI-powered data warehouse that lets data teams query unstructured, multimodal data - documents, images, web pages, video - with plain SQL. A UC Berkeley CS and Statistics double major, he previously led generative AI engineering at Snowflake and worked on the Skills and Knowledge Graph products at LinkedIn before founding Roe AI in 2023 to solve a problem that nagged him for a decade: the 80% of enterprise data that never gets used because it isn't in neat rows and columns.
Sammy Sidhu
Sammy Sidhu is the co-founder and CEO of Eventual, the company behind Daft, an open-source, Python-native data engine built to wrangle messy multimodal data like images, video, audio and LiDAR at petabyte scale. A UC Berkeley EECS grad with a background in high-performance computing and deep learning, he cut his teeth on Wall Street high-frequency trading and then on the perception stacks of self-driving cars at DeepScale (acquired by Tesla), Lyft Level 5 and Woven Planet (Toyota). After repeatedly watching brilliant engineers burn 80% of their time fighting data plumbing instead of building, he and Jay Chia turned that frustration into Eventual. Daft now processes petabytes daily at Amazon, CloudKitchens, Together AI and Essential AI, and the company has raised roughly $30M from investors including CRV, Felicis, Microsoft's M12 and Citi.