Niobia takes the guesswork out of some of tech's most expensive mistakes

San Francisco startup Niobia is tackling one of the costliest parts of battery and chip manufacturing: figuring out why defects happen. The company’s AI software deploys on production lines using self-contained GPUs mounted on equipment, where it scans for tiny issues such as pinhole-sized scratches. When agents spot a problem, they compile manufacturing data in one place and let engineers ask a chatbot what likely caused the failure—aiming to deliver answers in minutes rather than days. Founded by Gaurav Jha and Harveen Kaur, both with PhDs in chemistry, Niobia is also using open-source models to keep proprietary customer data off public AI clouds and offering encryption-key controls for security.