What is AMD’s AI Strategy for India?

Last month, to expand its research and engineering operations in India, AMD inaugurated its largest global design centre in Bengaluru. The AMD Technostar R&D campus is a key component of the company’s $400 million investment in India over the next five years. AMD recognises the potential India has for its global goals.

To understand AMD's strategy for India, AIM caught up with Andrew Dieckmann, CVP & GM, Data Center GPU, AMD; Brad McCredie, CVP at AMD; and Vamsi Boppana, SVP of AI at AMD, at the AMD Advancing AI event in Santa Clara to ask more about the expansion plans in India. “We are definitely focusing on that,” said Dieckmann. “We have a lot of our software team, AI team, and chip design teams in India and we definitely will continue building on that.”

He added that he sees India as a very competitive market since it is a very large market and expects his competitors to react. “We have a very strong roadmap, and we intend to lead in the market,” he added, talking about the launch of Instinct MI300X, and how it is going to compete with NVIDIA’s upcoming GH200, while AMD’s GPU is compared with H100s.

Dieckmann and McCredie highlighted how GH200 is going to involve a multi-chip module (MCM) design, which is something that AMD has been doing all this while, using GPUs and CPUs at the same time. 

“We're coming to market with a tried and tested technology,” McCredie said, adding that GPUs have always been used for generative AI. AMD is coming with a key differentiator in the market with world-class CPUs.

AMD to be a household name

In India, giants like Tata, Reliance, and Infosys have been partnering with NVIDIA to acquire its GPUs, and build generative AI in India. AMD also has similar undisclosed plans.

“Besides making the software easy to use and making compiling models easily, (one way) is giving access to the software and hardware,” said Boppana, talking about how it aims to take its momentum in India. “If I was a developer in India and asked how to use ROCm in India, it was not easy to get one as there were no ready cloud instances and the consumer cards were not supported for ROCm,” he explained.

“This has changed now.”

AMD has recently added support for ROCm for its consumer-based GPUs, such as Radeon and Ryzen. “This has increased access to our platforms and next year we are planning to add cloud instances for more access for programming on our Radeon GPUs which are easily available.” These developments will allow developers to come onto AMD’s platforms and build up in the country.

“With software development, and AI coming on PCs, I think AMD is right at the cusp of this innovation,” Boppana added, talking about developers in India. 

The easiest alternatives

“I think one of the reasons why AMD is going to lead from now is because of its partnerships,” said McCredie, highlighting partnerships with Microsoft, Meta, Oracle, and customers such as Databricks, Lamini, and Essential AI.

“The other thing is that the market definitely needs choice, and I think we have established ourselves as the most logical and easy to adopt alternative choice in the market,” he added about competing with NVIDIA. “The last point I'd point out is that we are coming to the market with something that offers better performance than the current state of the art technology from the incumbent”

AMD says that it is not comparing itself against last generation accelerators as the work their MI300X and other products perform are already better than the ones in the market. 

“Competition is good. It brings innovation,” said Mark Papermaster, CTO of AMD in an exclusive interaction with AIM. “It brings pricing that ensures value for the customers and spurs the industry forward. We have not only brought competition, but are also bringing in a leadership product in inference applications,” he further added about MI300X.

Focusing on the recent acquisition of Nod.ai, a software stack company that is now helping AMD, Papermaster said that AMD is always looking at the startup community in India. “We have a strong design presence in India. The country will, of course, be central for our AI product development, hardware and software product development efforts.”

Similarly, Gilles Garcia, senior director business lead, data centre communication group at AMD also told AIM in an interaction, “We are always monitoring what value we can bring and what value other companies can bring to us,” he said about investing in AI startups in India. 

“We have strong relationships with universities and we are also providing additional training to students. Then we bring them onto a very established internship programme,” he explained about how AMD has established an excellent pipeline for college graduate engineering in India. AMD’s AI strategy in India is from the ground up and from the top down as well.