Samay Maini is Co-Founder at Human Archive (YC W26), a multimodal data provider for robotics learning that collects and labels aligned data to develop dexterous and perceptive robots. The company deploys custom hardware in residential and manufacturing environments to deliver high-fidelity datasets to frontier robotics labs and foundation model companies.

Samay studied Electrical Engineering and Computer Science at UC Berkeley and worked at Sparkfish LLC, where he co-authored Augraphy, a Python library for document image augmentation. He also gained experience at Machine Learning at Berkeley (ML@B), IF Ventures, and Contoro Robotics, specializing in computer vision, NLP, and robotic perception.

Samay co-authored research papers published at ICDAR 2023 on Augraphy and ShabbyPages, demonstrating expertise in machine learning and document processing. His work on Augraphy has been cited by researchers at institutions including KU Leuven, University of Salerno, and Johns Hopkins University.

Human Archive raised $500K in convertible note funding from Y Combinator and is part of the Winter 2026 batch. The team dropped out of Stanford and Berkeley to move to Asia and collect the world's largest annotated multimodal dataset for robotics.

Human Archive serves frontier research labs and humanoid robotics companies developing foundation models for general-purpose robots. The company's infrastructure focuses on biomimicry research and delivers task-relevant, high-quality data to help reshape physical labor markets.