
For most of the last decade, India’s electronics industry meant assembly: importing components, printed circuit boards, displays, camera modules, chips, and putting them together into a finished phone, television, or appliance.
This activity is called Electronics Manufacturing Services, or EMS, and it became one of India’s genuine industrial successes.
Electronics production grew six times in a decade, from ₹1.9 lakh crore in FY15 to ₹11.3 lakh crore in FY25. Mobile phone manufacturing alone went from two plants in 2014 to more than 300 today.
But assembly sits at the lower-margin end of the supply chain. The company doing the assembling earns a thin return on each device; the larger part of the value sits upstream, in the components and the chips themselves.
For India’s electronics industry to grow beyond an assembly base, it needed to build backward into that value chain. That shift is now underway.

The clearest evidence of how far assembly had outpaced component manufacturing is a single figure: more than 90% of the printed circuit boards used in Indian electronic products are still imported, mostly from China, Taiwan, and South Korea.
A printed circuit board is not a minor part – it is the base almost every other electronic component sits on. For a country producing ₹11.3 lakh crore of electronics a year to import nearly all its PCBs says less about a weakness and more about how much of the value chain still sits outside the country.
This is the gap the Electronics Component Manufacturing Scheme is built to close.
A relationship that began with assembling Apple’s phones is now extending into manufacturing the components those phones are built from.
Components are one layer of backward integration. Semiconductors are a deeper one, and considerably harder to enter – chip manufacturing is more precise, more capital-intensive, and slower to qualify than assembling a finished device.
India’s entry point into this layer has not been full chip fabrication. It has been assembly, testing, and packaging, an activity the industry calls OSAT: a chip is fabricated on a silicon wafer inside a cleanroom, then cut apart, mounted, wire-bonded, and sealed into the package that ends up inside a phone or a car.
It is where India’s semiconductor mission chose to begin, because it needs less capital than a fabrication plant and can be brought online faster.

Micron, Sanand, Gujarat began partial production in early 2025, India’s first advanced semiconductor packaging line, backed by roughly $825 million of its total $2.75 billion India investment.
Kaynes Semicon, also in Sanand, built its OSAT facility from groundbreaking to commercial production in 14 months, a genuinely fast timeline for anything resembling a chip plant.
CG Power’s OSAT line, running since August 2025, already processes 500,000 chip units a day, with a roadmap toward 14.5 million a day at full capacity.
Fabrication itself, the stage before OSAT, is the layer India is only now entering. Tata Electronics planned facility in Dholera, Gujarat, is intended to be India’s first chip fabrication plant, and its commissioning date is the single most closely watched milestone in the country’s semiconductor programme.
Tata is separately building an OSAT facility in Assam, giving the same company a position at both the fabrication and packaging ends of the process.

Taiwan alone produces more than 60% of the world’s chips. That concentration is a large part of why the US CHIPS Act ($52 billion), the EU Chips Act, and India’s own Semiconductor Mission (₹76,000 crore) have all emerged within a similar window – every major economy is building some form of domestic capacity, and India’s programme is one part of that wider reconfiguration.
Fabrication and packaging are the manufacturing end of the semiconductor chain. Design sits earlier and separately – deciding what a chip does before it is fabricated – and is generally the highest-margin part of the industry, since it does not require owning a factory.
India’s Design Linked Incentive scheme backs this stage directly: a performance-linked incentive of 4 to 6% of net sales turnover for five years, capped at ₹30 crore per applicant, plus subsidised access to EDA design software and fabrication slots. Since 2022 it has committed ₹234 crore across 22 companies, against a total project cost of ₹690 crore.
InCore Semiconductors, incubated at IIT Madras, licenses its RISC-V processor cores, Azurite and Calcite, to other chipmakers rather than manufacturing chips itself, a moat built on validated IP rather than capital. It has raised $3 million from Peak XV Partners.
Mindgrove Technologies, also Chennai-based, had its V2600 SoC supported under the first round of DLI and has raised $10.35 million to date, including backing from Peak XV Partners and Surge, and has reached tape-out.
Both companies show the same pattern: real capital, real tape-outs, and an IP-first or licensing model that does not require owning a fabrication plant to compete.
None of these layers, components, OSAT, fabrication, design, were built in isolation. Behind them sits a set of decisions made across roughly the last eighteen months.
Private capital has started to move alongside the policy. In 2026, a coalition of Accel, Blume, Celesta, Premji Invest, Qualcomm Ventures, and Kalaari, with NVIDIA advising, committed more than a billion dollars to India’s deep-tech ecosystem, the first-time capital at this scale has aligned specifically with the semiconductor and EMS timeline.
The move from assembly toward components, packaging, and design is visible company by company. Grouping them by the layer they operate in makes the shift easier to see than any single company profile would.
Each company below is included for a specific reason: a genuine capital commitment, a manufacturing capability, or a technical position that is difficult for a competitor to simply replicate. Companies without a clear structural story of this kind are left out, rather than filled in for the sake of completeness.





Electronics is rarely the final product a customer buys. It is almost always a component inside someone else’s, which is what makes this shift matter well past its own sector.
Smart meters make the point concretely: 5.08 crore have been deployed from a near-zero base only a few years ago, and each one is an electronics product built on the same manufacturing base described throughout this piece.
Three other measures show the same shift from the EMS side of the industry, in exports and infrastructure rather than in chips.

KEY INSIGHT : The companies doing EMS work five years ago are, in several cases, the same companies now building components, packaging chips, or funding chip design, Kaynes and Tata Electronics among them.
What has changed is not the industry’s starting point, but the layer of the value chain it is choosing to compete in next, with policy, private capital, and individual company decisions now pointed in the same direction at the same time.
None of this replaces the assembly industry India already built. It sits on top of it, extending a manufacturing base that took a decade to reach ₹11.3 lakh crore into the components, packaging, and design layers that base did not previously include.