Advertisement

Artificial intelligence is transitioning from promise to reality, with memory and storage as the primary bottleneck in handling data at speed. To unlock AI’s promise, businesses must embrace a data-centric approach and leverage the right technology foundation.  In the case of Generative AI training and applications, Micron’s industry-leading technologies enable the latest generation of faster, more intelligent global infrastructures that make AI model training, machine learning training and generative AI solutions possible.

In this episode of Micron presents The Power of Memory on CNBC-TV18, Micron leaders lay out how AI is designed, deployed, and scaled across the company, and how India is instrumental in that effort.

Why memory drives modern AI

“AI can be thought of in layers,” says Samir Mittal, Corporate VP, AI for Silicon and Systems Design. “The first layer helps us find new physics, new transistors, new devices. The second layer is design space optimization. The third, now most popular, is reasoning, where we can instruct AI to plan and even build the plan itself.” In every layer, the constraint is data movement. As Mittal puts it, “In all of these, AI computing is memory centric. You need lots of memory and high-bandwidth memory to enable new computing paradigms.” The result is faster discovery, tighter design loops, and shorter time to market. “The possibilities with AI are literally quite endless.”

Smart manufacturing, measurable gains

Micron’s AI journey did not begin with the recent wave of generative tools. “We started the smart manufacturing AI organization back in 2015,” says Koen De Backer, CVP, Smart Manufacturing and AI. “Our factories are highly automated, they produce a lot of data, and we leverage that for machine learning models, from auto diagnostics to image analytics on the line.” The goals are throughput, yield, and quality. “Everything we can do to produce more wafers with the same assets is a big driver. We use simulation and scheduling optimization, we focus on faster yield ramps for the latest products, and we protect quality end to end.”

Generative AI is now embedded in daily work. “AI code assistance is a massive productivity gain,” De Backer says. “India is our largest location in headcount for smart manufacturing. Data scientists, ML engineers, solution architects, and operations teams in India help drive the AI transformation across front-end, assembly and test, engineering, and business functions.”

IT as a proving ground for AI at scale

AI is also changing the employee experience. “Our GenAI chatbot around IT ticketing used to require a help-desk call for something like a password reset,” says Anand Bahl, SVP and Chief Information Officer. “Now it is automated. What took a phone call and maybe hours can be done in five minutes.” The same pattern shows up in procurement. “Approving purchase orders meant diving into multiple systems. You can now do that automatically inside the tools we have created.”

The payoff is quantifiable. “For custom developers, we have seen a 20 to 30 percent efficiency gain from code generation,” Bahl says. “Microsoft Copilot helps with meeting prep and long email summarization. It is a time saver when you are trying to ingest large amounts of information.”

MIRC- Enabling silicon design and research

The Micron India Research Center (MIRC) concentrates on two tracks, says Kiran Gunnam, Director, Automated Layout, CoPilot. “We are infusing AI CoPilot into our engineering process, especially in silicon development, and we are building large-scale distributed training systems that include HBM and enterprise SSDs.” CoPilot helps engineers navigate and explain large code repositories, highlight issues during runs, and accelerate onboarding to complex flows.
Gunnam notes a second benefit in technical research. “LLMs and reasoning agents can summarize heavy paper loads so experts know where to focus. Retrieval-augmented flows let you ask questions over internal documents without weeks of bespoke scripting. It is a general tool that keeps engineers focused on insight rather than plumbing.”

Products for AI, culture for speed

Micron builds what AI needs to run. “Think of GPUs paired with high-performance memory,” Mittal says. “Without a fast place to store and access data, the premise of AI falls apart.” Anand Ramamoorthy, VP and Managing Director, Micron India, connects that to local impact. “A lot of those products are designed in India in conjunction with other sites. Almost every engineering employee at Micron India plays a fair share in enabling this AI revolution.”

Ramamoorthy frames the moment as a mindset shift. “We have a unique chance to create a culture of being AI first by treating AI tools as a default primary option,” he says. “. “Our teams are pulling AI into layout verification, validation, and debug, building internal tools and chatbots that spot anomalies sooner, and making the site more productive and more resilient.”

The throughline

Discovery, design, manufacturing, IT, and daily workflows now share a common pattern. AI thrives when data moves quickly and predictably. Memory and storage set that tempo. Micron pairs the engines that power AI with the practices that operationalize it, while India supplies both scale and depth of talent to keep the flywheel turning.

The pages slugged ‘Brand Connect’ are equivalent to advertisements and are not written and produced by Forbes India journalists.

First Published: Nov 29, 2025, 19:07

Subscribe Now
Advertisement
Advertisement

What others are reading

Advertisement

Latest News