SwitchOn raises $8 Mn to scale AI manufacturing inspection

Key Takeaways

• SwitchOn raised $8 million in a pre-Series B round led by IvyCap Ventures, bringing total funding to $13.3 million
• The startup's DeepInspect platform detects defects at sub-150-micron precision, processing over 1,200 products per minute
• Physical AI startups are attracting investor attention as manufacturers adopt vision systems for quality control automation

SwitchOn, a Bengaluru-based startup building AI-powered visual inspection systems for manufacturing, has raised $8 million in a pre-Series B funding round. IvyCap Ventures led the round, with SIG Tattva and Trifecta Capital participating. The company will use the capital for international expansion and R&D in physical AI.

The round comprised a mix of equity and debt, though SwitchOn declined to specify the breakdown. Including this raise, the seven-year-old startup has now raised $13.3 million.

What does SwitchOn's inspection technology actually do?

Founded in 2017 by Aniruddha Banerjee and Avra Banerjee, SwitchOn builds computer vision systems that spot defective products on factory production lines. The technology replaces or augments human quality inspectors who manually check products, a process that's slower and prone to fatigue-related errors.

The company's flagship platform, DeepInspect, runs on edge devices directly on the production floor rather than in the cloud. SwitchOn claims it can detect surface defects as small as 150 microns, processing more than 1,200 products per minute. For context, 150 microns is roughly the width of a human hair.

SwitchOn says the platform can cut quality-related costs by up to 50%. That's a bold claim, but the customer list suggests traction: Unilever, Bosch, Maruti Suzuki, and ALPA all use the technology. The startup reports deployments across 170 production lines in over 60 manufacturing facilities on four continents.

Why investors are watching physical AI

This funding reflects a broader pattern. Investors are increasingly interested in startups that move AI from software into the physical world, combining machine learning with cameras, sensors, and industrial equipment.

The logic is straightforward. Manufacturing quality control is expensive, error-prone, and scales poorly with human labor. A trained AI system running on cameras can work 24 hours without fatigue, catch defects human eyes miss, and provide data trails for compliance. As AI models improve and edge computing gets cheaper, the economics keep shifting toward automation.

SwitchOn isn't alone in riding this wave. Earlier this month, robotics startup Mowito raised $3 million in pre-seed funding to build AI that trains industrial robots by watching human operators. The startup already has deployments at a Fortune 500 automotive company and a major electronics contract manufacturer.

Y Combinator-backed Human Archive also recently raised $8.2 million to build what it calls the largest human sensorimotor dataset. These datasets train robots and manufacturing AI systems to understand how humans move and interact with physical objects.

Where SwitchOn goes from here

The company said it will use the funding for three priorities: international expansion, R&D in next-generation physical AI, and scaling go-to-market capabilities across manufacturing sectors.

SwitchOn serves industries including FMCG, electronics, automotive, and pharmaceuticals. Each vertical has different inspection requirements, from checking pill coatings in pharma to spotting paint defects on car bodies. The ability to customize models for different use cases will likely determine how quickly the company can expand.

The startup's previous round was a $4.2 million Series A in 2023, led by Axilor Ventures and pi Ventures. The jump to an $8 million pre-Series B suggests investors are seeing enough commercial traction to double down.

The market opportunity

The global machine vision market in manufacturing is estimated at $2.3 billion in 2024, with projections suggesting it could reach $4.2 billion by 2030. Those numbers explain why investors are willing to fund multiple players in the space. There's room for startups to carve out niches by vertical, geography, or technical approach.

For Indian startups specifically, the Make in India initiative creates domestic demand as manufacturers upgrade facilities to compete globally. SwitchOn's international expansion plans suggest it sees domestic traction as a springboard rather than the end goal.

Frequently Asked Questions

What is SwitchOn and what does it do?

SwitchOn is a Bengaluru-based startup that builds AI-powered visual inspection systems for manufacturers. Its DeepInspect platform uses computer vision to automatically detect defective products on production lines, replacing or supplementing human quality inspectors.

How much has SwitchOn raised in total?

Including the latest $8 million pre-Series B round, SwitchOn has raised approximately $13.3 million to date. The company previously raised $4.2 million in a Series A round in 2023.

What industries does SwitchOn serve?

SwitchOn serves manufacturers across FMCG, electronics, automotive, and pharmaceutical sectors. Its customers include Unilever, Bosch, Maruti Suzuki, and ALPA.

What is physical AI in manufacturing?

Physical AI refers to artificial intelligence systems that operate in the real world rather than purely in software. In manufacturing, this typically means AI combined with cameras, sensors, and industrial equipment to automate tasks like quality inspection, robot control, and production monitoring.

Source: Inc42 Media / Palak Sharma