Homegrown Indian Startups Target Specialised AI Chips for Everyday Devices

Homegrown Indian startups are developing specialised artificial intelligence chips for cameras, cars and other connected devices as the country enters the edge AI sector. Pratik Parmar of startup Atomo displayed a circuit board that lets cameras and machines spot fires or track people locally without sending data to the cloud.

Growth and Market Potential

India's semiconductor market is projected to grow from 64 billion US dollars in 2026 to 200 billion US dollars by 2035. Industry leaders note that rather than building the world's fastest AI processor, focusing on intelligence for everyday electronics offers a more realistic path. A government chip-design programme has already helped 105 startups access design tools and produce about 35 chip designs ready for manufacturing.

Advantages of Local Processing

Edge AI chips keep sensitive data on-site, reduce cloud costs and work when internet connectivity fails. Atomo has sent circuit boards for testing in Japan, South Korea and the Middle East. Meanwhile, BigEndian Semiconductors founder Sunil Kumar noted that India's large scale makes it an ideal testing ground for handling massive events like the Kumbh Mela. Investor backing has also grown, with Indian edge AI startups raising 17 million US dollars in 2026, up from 4.27 million US dollars in 2022.

Context

Edge AI chips process data locally on devices rather than relying on distant data centres, providing faster response times for applications like car safety systems and factory cameras. India hosts a significant share of the world's semiconductor design engineers, giving domestic firms a strong talent base to build specialised hardware.