India’s chip startups struggle to fund leap from design to commercial production

The Tracxn data shows that the funding funnel narrows after seed. From 2023 through 2026 so far, the sector recorded 94 seed and angel deals against just 20 Series A deals. Seed and angel funding stood at $44.1 million across 32 deals in 2024, $33.4 million across 28 deals in 2025 and $19.2 million across 15 deals this year.

Series A activity has been thinner, though improving, with four deals worth $25 million in 2024, six worth $53.6 million in 2025 and eight worth $72.9 million this year.

Growth-stage funding reached $423.1 million in 2025, with ILJIN Electronics and Tessolve Semiconductor accounting for $348.4 million of the total. ILJIN is an electronics manufacturing services company, while Tessolve provides semiconductor engineering, testing and validation services. Neither is a young fabless startup that designs and sells its own chips.

State support has been strongest at the design stage. At least 24 projects are being supported under the Design Linked Incentive scheme, while 105 companies have received access to advanced chip-design tools, the government said in June. It said 23 chip designs had completed tape-out - stage at which a finished design is sent to a foundry for manufacturing.

Tape-out is a critical milestone because it determines whether years of engineering work can be turned into physical silicon. But reaching this stage does not mean a startup has a product ready for commercial sale. The chip must still return from the foundry, undergo packaging and testing, work reliably with software and hardware, and be qualified by customers.

Splitting the ask

A Bengaluru-based founder of an AI processor startup, speaking on condition of anonymity, said the money often runs out before the first silicon is produced. The startup has completed its design and tested it in simulations and on a programmable prototype, but needs capital to turn that into a physical chip.

The founder said an initial batch of about 100 chips would cost roughly $6.3 million. That would only establish whether the design works after manufacturing, how many chips are usable, and whether packaging and software hold up. Unable to raise the full amount, the company has broken the requirement into stages.

“We initially sought $40-50 million to take the product through tape-out, early production and its next version, but now we have split that requirement into smaller rounds, starting with $15 million. That should fund tape-out and an initial run of about 1,000 chips,” the founder said.

Investors are often reluctant to commit large sums before a chip has been taped out.

An investment banker who advises deep-tech companies said that investors assess whether a pre-tape-out design can be manufactured reliably and whether it has a market. The larger the cheque, the more founders must demonstrate a credible development plan, potential customers and a route to market.

“Founders may need to raise smaller initial rounds, meet milestones and return for more capital in a staggered manner,” the banker said.

Why funds hesitate

A senior partner at an early-stage venture capital firm said the arithmetic of a frontier chip company is difficult for a conventional venture fund to approve because the capital requirement arrives long before evidence of commercial success.

Companies attempting to compete at scale with global chipmakers may need $500-750 million, this person said, adding that such patient capital sits outside the current Indian venture model.

Hemant Mohapatra, partner at Lightspeed and a former AMD executive, said the chip market is divided between capital-intensive frontier AI-chip companies and makers of lower-cost chips for appliances, electric vehicles, etc.

“A first tape-out at a leading node can cost $30-40 million, while a tape-out on a mature 160nm process may cost $150,000-200,000,” he said.

This makes it harder for startups that lack either globally competitive technology or a clear path to high-volume demand to raise funds. “It’s fairly bimodal right now. If you’re sitting in the middle, I think it’s much harder for you to prove your case,” Mohapatra said.

For mass-market chip companies, investors want evidence of customer demand alongside technical validation. “The bigger question is whether you have an anchor customer who will buy a million chips the moment you bring it out,” he said.

After the foundry

The financing challenge does not end once a chip design reaches a foundry. The chip must be packaged, tested and integrated into a customer’s product. This work is often handled by outsourced semiconductor assembly and test, or OSAT, providers, which package bare silicon so it can be mounted on circuit boards and test whether it works reliably.

Without access to these services, a startup can receive chips from a foundry but still be unable to sell a finished product. As production grows, it must also check manufacturing yields, qualify components and materials, and secure backup suppliers.

Shashwath T.R., co-founder and chief executive of Mindgrove Technologies, said companies must reserve foundry and OSAT capacity before demand is certain. “We have to give forward projection to the foundry and to the OSAT to be able to reserve that capacity in advance,” he said.

Calligo Technologies, which says it manufactured an eight-core processor at Taiwan Semiconductor Manufacturing Co.’s 28nm node, raised about $1 million in seed capital and is in talks for $12-15 million for a 64-core successor.

“Once the semiconductor samples are manufactured and received, you cannot make any changes to the silicon if there are bugs in the design,” said Anantha P. Kinnal, its co-founder and managing director. “Bugs have to be fixed and design has to go through manufacturing again.”