Computing’s Top 30: Amod Agrawal

Amod Agrawal 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’m Amod Agrawal, and I work as an Applied Scientist at Amazon Lab126, the organization behind Amazon’s device ecosystem, including Alexa, Echo, Fire TV, Kindle, and other connected devices. I develop sensing algorithms that enable everyday devices to perceive spatial context, allowing them to understand where people and devices are and how they move through physical environments. I specialize in prototyping these technologies in real-world environments and translating them into systems that operate reliably across millions of devices. My work sits at the intersection of wireless sensing, indoor localization, and edge computing, enabling intelligent environments where AI assistants respond naturally to human presence and behavior.

Prior to joining Amazon, I was a graduate student at the University of Illinois Urbana-Champaign, where I researched how everyday invisible wireless signals can reveal far more about our environment than most people realize. It's a niche area that doesn't always get the spotlight, but the applications are quietly everywhere.

What do you consider your highest achievement so far?

I have led the development of large-scale sensing systems that operate in real consumer environments. While much of the research is often evaluated in controlled settings, my work has focused on translating these ideas into reliable systems that function across 600 million Alexa devices. Real-world environments introduce noise, interference, device diversity, and unpredictable user behavior. Designing algorithms that remain robust under these conditions requires careful engineering, extensive experimentation, and validation at scale. Being able to bridge that gap between research concepts and production systems has been very rewarding.

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

Looking ahead, I am expanding the role of wireless sensing in enabling ambient intelligence. Today, many computing systems rely heavily on explicit user input or privacy-intrusive approaches such as camera-based tracking. AI systems are only as effective as the context they receive, and our physical environments will play a much larger role in providing that context in the future. Devices will be able to understand presence, activity, and spatial relationships, and use that information to enable more seamless and intuitive interactions.

Wireless signals are uniquely well positioned to enable this form of intelligence because they are already embedded in everyday devices through technologies such as Wi-Fi and Bluetooth. By developing robust sensing frameworks, we can unlock new capabilities without requiring entirely new sensing infrastructure. Ultimately, my work builds systems where the physical environment itself becomes an intelligent interface between people and technology.

How are you currently involved in the tech community aside from your job?

I am passionate about engaging with the research and engineering community and contributing beyond my core job responsibilities. I regularly participate in academic conferences and workshops related to wireless sensing and mobile computing. These venues provide valuable opportunities for researchers and industry practitioners to exchange ideas. Academic researchers often introduce breakthrough concepts, and interacting with them brings fresh perspectives to my work, while sharing industry considerations helps ground research in real-world challenges. I actively contribute to trade publications and online articles where I share perspectives on emerging research and industry trends.

I also serve on the technical program committees for research conferences and academic research grants. These roles allow me to help evaluate emerging research directions and support the broader research community.

In addition, I enjoy mentoring students and early-career engineers interested in sensing systems and applied AI. In academia, research is often optimized for benchmark performance, whereas industry systems must solve the majority of real-world cases with high reliability. Helping others navigate this transition from academic research to real-world system development is particularly rewarding for me.

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

One important development is the convergence of edge devices and generative AI. Recent advances in AI are making it possible to develop agents with reasoning capabilities. At the same time, the rapid growth of smart home devices, smartphones, and AI wearables is creating dense networks of systems that can capture rich information about physical spaces. Most large models today are still effectively blind to the physical world. They operate on text and images but lack access to the sensor-rich context that ambient systems can provide. Bridging this gap is one of the most interesting open problems in the field today.

In our community, this vision is often described as ambient intelligence, where homes, vehicles, and everyday devices can automatically adapt to human behavior, collaborate to understand context, and provide interactions that feel more natural. I am enabling this through privacy-preserving sensing approaches that leverage invisible signals in the environment instead of relying on cameras.

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

As homes, vehicles, and everyday infrastructure become smarter, these systems can evolve beyond convenience and begin supporting critical life functions. For example, falls remain one of the leading causes of injury-related death among older adults. Sensing systems can help detect falls, monitor gait, track sleep quality, and even identify conditions such as sleep apnea. These capabilities could play a significant role in healthcare and elder care over the next several years.

Similar ideas extend to other environments such as vehicles, where sensing technologies can monitor driver attention and behavior to improve road safety. Earlier in my career, I contributed to Project HAMS at Microsoft Research India, where we developed a platform that detects unsafe driving behavior using only a smartphone. The system was adopted by the Government of India and deployed across dozens of Regional Transport Offices, helping automate behind-the-wheel driving tests for hundreds of thousands of drivers. Such innovations can directly shape governance and public policy at scale.

More broadly, AI has the potential to expand access to technology and lower the barrier to contribute to society. As AI tools become more capable at executing complex tasks, the ability to apply creativity and problem-solving to real-world challenges becomes increasingly important. This shift has the potential to level the playing field and enable more people to build solutions that address societal needs.

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

Working on zero-to-one ideas quickly teaches you that every aspect of product development matters. I have had the opportunity to work on several such projects where success required engaging with many cross-disciplinary aspects of building a system. One of the most important lessons from those experiences is that the way a problem is framed is often more important than the solution itself.

Engineers solve the software and systems challenges, scientists focus on algorithms and modeling, designers think about how something should be built and for whom, and user experience researchers work directly with people to understand their needs and gather feedback. Having worked across these perspectives, I have learned that meaningful innovation often emerges when these viewpoints are considered together. It may slow the process initially, but it helps prevent building technically impressive systems that ultimately do not solve real problems.

What advice would you give to young professionals entering your field?

My main advice is to focus on building strong fundamentals while also developing the ability to work across disciplines. While AI is becoming a powerful tool for solving complex problems, the professionals who shape its impact are those with strong domain knowledge. Early in the career, it is often beneficial to avoid narrowing oneself too quickly into a single specialization. In my own work, ambient intelligence sits at the intersection of signal processing, machine learning, applied AI, wireless networking, and embedded systems. Driving impact in this space requires a strong foundation across all of these disciplines.

It is also important to become comfortable working with ambiguity. Research-to-product roles rarely come with clearly defined specifications. Often the challenge is not just building a solution but defining the problem statement itself. The people who thrive in these environments are those who can ask thoughtful questions, iterate quickly with imperfect information, and know when to explore an idea deeply versus when to move on.

I also encourage young professionals to publish their work regularly, even when your ideas may feel incomplete. The exchange of ideas has a compounding effect, and many small contributions collectively move the field forward. Finally, technical ability is only one part of a successful career. Being reliable, collaborative, and easy to work with is equally important. Technical skill opens doors, but character and teamwork sustain long-term impact.

Stay Connected

Find Amod on the following platforms:

• Github
• LinkedIn

Building Your Influence and Legacy

While the IEEE Computer Society is discontinuing the Top 30 Early Career Professionals award, emerging professionals still have ample opportunity to establish an influential legacy and gain industry recognition through active participation and the Society's resources.

You can make a significant impact by:

• Supporting Conferences: Volunteer your time and expertise.
• Developing Knowledge: Become a peer-reviewer to see the inner workings of publications and help maintain the quality of research presented to the world
• Mentoring: Become a mentor to guide and support others.
• Leading Local Activities: Coordinate events and activities for local chapters.

Help shape your community today! Explore various opportunities and discover volunteer positions with the IEEE Computer Society to get involved.