Tech News, Blockchain, Cryptocurrency and the Internet

Marvell to invest $250 million in India and double semiconductor workforce

AI chip and Bengaluru engineering campus representing Marvell's $250 million India R&D investment
Original illustration of Marvell’s planned $250 million India investment in AI chip and data-infrastructure R&D.

US chip designer Marvell Technology will invest $250 million (about ₹2,090 crore) in India over the next three years, expanding its Bengaluru engineering facility and increasing its local workforce. The announcement, made on July 29, 2026, puts India more firmly inside the research-and-development network behind the chips used in AI systems, cloud infrastructure and high-speed data centres.

Marvell said the investment will cover technology development, talent and infrastructure. Reuters reported that the company also plans to double its India headcount. The expansion is important, but it should not be confused with a semiconductor fabrication plant: Marvell is a fabless chip company, so its Indian operations are primarily focused on design, engineering and related R&D rather than manufacturing silicon wafers.

What Marvell plans to build in India

The centrepiece is an expansion of Marvell’s Bengaluru facility. According to the company’s official announcement, the new investment will support development of next-generation technology for AI and data infrastructure. Marvell has operated in India for roughly two decades and also has engineering teams in Hyderabad and Pune.

The precise hiring timetable and eventual number of new employees were not disclosed in the announcement. However, Reuters reported that Marvell intends to double its workforce. In November 2025, the publication said the company had about 1,700 employees in India, with work spread across Bengaluru, Hyderabad and Pune. The eventual total may therefore be substantial, although readers should wait for updated company disclosures before treating a specific number as final.

Why the investment matters

AI infrastructure does not depend on graphics processors alone. Modern data centres also require networking silicon, optical connectivity, storage controllers and custom processors to move huge volumes of data efficiently. These are areas where Marvell competes, which makes its India expansion relevant to the broader AI hardware supply chain rather than just the consumer-device market.

India has a deep pool of chip-design talent, but its domestic semiconductor ecosystem is still building scale across manufacturing, packaging and testing. A larger Marvell engineering presence can add high-value design jobs and give Indian teams more responsibility for products used globally. It also complements the wave of investment in Indian AI infrastructure, including HCLTech’s planned AI data centre in Odisha.

For Marvell, the move is also a capacity decision. The company’s fiscal 2026 proxy filing said data-centre revenue had grown to more than $6 billion and represented roughly three-quarters of total revenue. Expanding engineering resources in India gives it a larger base from which to address demand from cloud operators and AI infrastructure customers.

What the announcement does not mean

The $250 million figure is an investment commitment over three years, not revenue generated in India or the cost of a new fabrication facility. Marvell did not announce a wafer fab, a production start date or manufacturing capacity. That distinction matters because chip design centres require far less capital than leading-edge fabrication plants, which can cost many billions of dollars.

Even so, design and product engineering are critical parts of the semiconductor value chain. If Marvell follows through on the facility expansion and workforce growth, India will gain a larger role in developing the networking and compute infrastructure behind AI services used worldwide.

Sources

Share this article
Shareable URL
Prev Post

How to upload and analyse PDFs in ChatGPT without missing information

Leave a Reply

Your email address will not be published. Required fields are marked *

Read next