As semiconductor technology advances to the atomic scale, the complexity of devices increases. Chip manufacturers face challenges such as defects and variability, stemming from new transistor architectures and multi-chip integration using advanced packaging techniques. KLA’s process control products are essential for identifying, sourcing and resolving production issues, and for KLA’s artificial intelligence engineers, this means partnering with product teams across the company to develop AI-driven solutions that help customers solve their most difficult challenges, improving yield and chip performance, and accelerating the next wave of semiconductor innovation.

According to Pradeep Ramachandran, director of KLA’s AI and Advanced Computing Lab in Chennai, India, it’s also an exciting reminder that the path he started down over 20 years ago remains central to that innovation.

“Every AI project we tackle is an opportunity to dive deep into the how and why,” Pradeep explains. “It’s not just about making things work — it’s about understanding the underlying physical mechanisms, understanding when and how a defect can occur during manufacturing. It’s a level of curiosity and precision that sets us apart from AI application development in other industries.”

From Lab to Fab: Solutions that Span from Software Development to Hardware Development

The AI and Advanced Computing Lab (AI-ACL) team, which includes engineers in India, Israel and California, engages with KLA’s product teams to capture and adapt recent developments in AI and high-performance computing to the company’s systems and technologies. For years, their work has touched multiple product lines.

It’s tightly knit work called co-optimization that focuses on the interplay between software and the impacts on the hardware. Essentially, what is envisioned and developed in a lab often can look much different when deployed on a tool. Pradeep’s team considers and designs around the conditions and limitations of AI algorithms operating on hardware in a semiconductor manufacturing environment.

“By developing advanced AI algorithms in KLA systems, we are helping semiconductor manufacturers advance chip technologies, particularly for chips used for AI applications.”

The Artificial Intelligence Flywheel: Helping Make AI Chips that Drive more Powerful AI Applications

“The expanding use of AI across industries is driving the need for AI chips in data centers and edge devices,” Pradeep explains. “These chips are often complex packages containing multiple chips integrated using advanced techniques. KLA’s products that support leading-edge logic and memory manufacturing and advanced packaging are a significant area of growth for us. We use AI to help our customers make the next-generation AI chips that ultimately help improve our systems’ AI performance.”

A Journey in Artificial Intelligence (AI)

Pradeep Ramachandran’s adventure in artificial intelligence began during his undergraduate years at the Indian Institute of Technology when he took a class called reinforcement learning, a field of artificial intelligence used for robotics. After earning his Master of Science and Ph.D. from the University of Illinois Urbana-Champaign, Pradeep took that curiosity to work on leading-edge technologies at various companies. In 2021, he joined KLA to build and lead the AI and Advanced Computing Lab.

“The technical depths are unmatched by my previous jobs – the complexity of what our teams deal with is fascinating”

KLA’s Global Approach to Nanoscale Challenges

Halfway around the world, another team of software engineers, algorithm engineers and physicists at KLA’s AI and Modeling Center of Excellence in Ann Arbor, Michigan brings a complementary set of skills and knowledge to the table, building physics-based models and solutions that ultimately improve image processing capabilities – a critical differentiator in many of KLA’s inspection and review, metrology and data analytic product lines that detect and categorize defects and identify variations during production of extremely complex chips.

“In many ways, KLA is an AI company, especially as it pertains to image processing. We’re very good at applying AI to image processing and physics,” explains Ann Arbor AI and Modeling Center of Excellence leader Vijay Ramachandran. “We pioneered using and incorporating AI into our products over a decade ago and have redesigned our compute architecture to incorporate GPUs for more robust algorithmic capability, which greatly enhances our systems and helps improve overall performance.”

KLA invests heavily in R&D to create these unique technologies; hardware and software subsystems that enable industry-leading performance. KLA’s hardware – light sources, optics, sensors, stages – is optimized to find and identify nanoscale defects and variations, which in turn produces massive amounts of image and system data. Operating at the forefront of algorithm development, KLA’s data scientists and engineers use innovative physics and AI-based algorithms to transform these challenging data sets into meaningful and actionable data.

“What KLA tools deliver,” Vijay says, “is information, and it’s key to our customers controlling their semiconductor manufacturing processes and helping improve their quality and yield.”

A Career Formed at the Intersection of Hardware and Software Expertise

Vijay Ramachandran came to KLA after earning his master’s degree, and as a software engineer, Vijay’s initial work included developing machine-control software for robot arms and lens manipulators. But he was fascinated by KLA’s technologies such as broadband plasma (BBP) patterned wafer inspection systems and, especially, machine learning – a concept that ultimately led him to his current pursuit at the intersection of the physical and digital worlds.

“I see my role as helping our algorithm, software and physics specialists connect the threads between functions,” Vijay says. “These teams have a fundamental impact on multiple product lines. Looking at these functions holistically is a key part of building efficient and robust systems and achieving high performance in our products.”

The Human Intelligence Behind the Artificial Kind

KLA’s experts in physics modeling, AI, algorithms, software and compute hardware uncover solutions hidden in large data sets and scale our solutions to support discovery at the edge of the nanoverse. Interested in exploring a unique KLA career in Artificial Intelligence?