Computing’s Top 30: Aditya Atluri

Aditya Atluri is one of our "Computing's Top 30 Early Career Professionals" for 2025. This program seeks to highlight an esteemed group of rising stars who earned this honor for their exceptional early-career achievements and role in driving advancements across the computing landscape. 

Introduction

I am Aditya Atluri, and I work as a Performance Architect at NVIDIA. My primary focus is on Hardware-Software Co-design for NVIDIA GPUs built for AI, encompassing architectures like Hopper, Blackwell, Rubin, and our next-generation designs.

What inspired you to pursue a career in technology?

Ever since I was a kid, I have been deeply fascinated by microprocessors. I was always curious about the underlying mechanics—how the different components of a processor seamlessly communicate and function together to power complex systems. That childhood curiosity naturally evolved into a passion for computing architecture.

What do you consider your highest achievement so far?

I’ve been fortunate to work on a few highly impactful projects. During my time at AMD, I co-authored the HIP runtime API, which is now a foundational layer for AI applications running on AMD GPUs, and I created RCCL, which is the AMD implementation of the NCCL API. At NVIDIA, my proudest achievement has been contributing to the definition of the Hopper architecture and its CUDA software ecosystem, optimizing hardware-software synergy to drive unprecedented performance for AI workloads.

How do you plan to continue or build on that success?

Having played a role in advancing GPU performance for AI at both the hardware and software levels, I want to broaden my scope. Moving forward, I am focused on solving larger, system-level challenges rather than just chip-level optimizations. My goal is to drive end-to-end performance improvements across the entire pipeline for AI inference and training, ensuring that our infrastructure can scale efficiently as models grow.

Who do you draw inspiration from and how did that motivate you in your education or career?

I have always been highly self-motivated and was a "GPU fanatic" long before the AI boom made them mainstream! However, I drew a lot of direction from industry pioneers and academia. Early on, I closely followed the work of leading PhD students and professors across the U.S. By studying their research and actively networking with them, I was able to mold my career path and align myself with the people and projects that were pushing the boundaries of technology.

How are you currently involved in the tech community aside from your job (volunteering, open-source projects, mentoring, etc)?

In my past roles, I was heavily involved in open-source projects, such as creating RCCL. Currently, due to the highly proprietary nature of hardware-software co-design at NVIDIA, my direct involvement in external technical communities is naturally limited to protect IP. However, I consider my primary contribution to the community to be foundational: building the underlying compute infrastructure and performance frameworks that empower developers, researchers, and open-source creators worldwide to build the next generation of AI.

Is there any emerging technology or industry segment you find exciting or interesting?

I am incredibly excited about AI agents and large-scale systems. Agents are going to fundamentally transform how we interact with the internet, data, and each other. I truly believe that the AI infrastructure we are building today is a modern engineering marvel, arguably on the scale of building the Pyramids. The sheer concentration of the world's smartest minds collaborating on this problem is astonishing. When you factor in the massive leverage we have through simulation and high-speed computation, the scale of what we are constructing is unprecedented.

How do you see technology shaping humanitarian efforts or social good in the next 5 years?

In humanitarian logistics, AI will dramatically improve our ability to monitor crises, deploy resources, and measure impact far faster and more accurately than ever before. On the social side, while there are always challenges with digital safety, AI is becoming deeply integrated into platforms to help filter harmful content and protect users. On a more personal level, AI is an incredible tool for bridging cultural and linguistic gaps. As a non-native English speaker, I’ve seen firsthand how AI can help refine communication, steer nuances, and foster better global collaboration.

If you have ever worked cross-discipline, how did that influence your way of thinking or the way you approach your work?

I believe the days of building an impactful career entirely within a single silo are fading. For example, AI researchers couldn't just stick to model architecture; they had to pivot to incorporating reinforcement learning to give these models better reasoning capabilities. You have to be adaptable. Approaching problems with a cross-disciplinary mindset is the best way to find the most efficient and elegant solutions. This is the very essence of hardware-software co-design: evaluating a bottleneck from multiple domains to determine whether it is best solved in silicon, in the software stack, or somewhere in between.

What advice would you give to young professionals or recent graduates who are trying to enter your field?

Stay adaptable and be okay with change. Try not to become emotionally invested in a single solution, technology, or job title. Instead, focus on finding interesting, high-impact problems that the market actually values and needs solved. If you haven't found that problem yet, it just means you need to keep exploring and learning.

Stay Connected

Connect with Aditya Avinash Atluri, on LinkedIn.

Dig Into Resources

Explore the various career guides, curricula, white papers, and technical insights and other content produced by IEEE Computer Society organized by topics in computing.