Introduction
The increasing complexity of machine learning systems demands a new approach to their design, optimization, and deployment. While tremendous progress has been made in ML algorithms and hardware acceleration, building reliable, efficient, and scalable ML systems requires expertise that bridges traditional disciplinary boundaries. The evolution of computing offers a valuable historical parallel: just as Computer Engineering emerged as an engineering discipline in the 1970s to bridge Electrical Engineering and Computer Science, there is a clear and growing need for Machine Learning Systems Engineering (MLSE) to unite ML algorithms, systems design, and practical implementation. This integration is essential for addressing challenges from hardware acceleration to deployment frameworks and operational infrastructure.
However, establishing this new discipline requires more than technical innovation—it needs structured programs to identify and nurture emerging talent. The EECS Rising Stars program has shown how targeted initiatives can strengthen the academic pipeline by identifying outstanding students and facilitating vital connections within academia. Inspired by this successful model, we started the “ML & Systems Rising Stars” program to build a similar ecosystem for young researchers working at the intersection of ML and systems.
The ML & Systems Rising Stars program, established in 2023, recognizes outstanding researchers who can navigate both the algorithmic complexity of ML and the systems challenges of implementing these algorithms efficiently and reliably. Now entering its third year, the program has become an important initiative for building a diverse and collaborative community of next-generation leaders in ML systems research.
As we prepare for the 2025 workshop, to be hosted by Meta in May 2025, we share our experiences building this program and invite the SIGARCH community to engage with this initiative—whether through nominating promising students, participating as mentors, or collaborating on future program development.
Building a Community of ML Systems Researchers
The ML & Systems Rising Stars program approaches community building through a carefully designed structure. Each year, we select 30-40 Rising Stars (2024; 2023) through a process that looks beyond technical excellence to identify researchers who can help advance the field in new directions. Our program committee, comprising members from academia and industry, reviews applications with attention to candidates who can bridge different aspects of ML systems work. This approach has proven particularly effective in identifying researchers who might otherwise work in relative isolation within their subfields.
The selection process is competitive yet diverse and inclusive. We have an extensive outreach strategy to ensure we reach a broad pool of candidates. Our announcements are distributed through university mailing lists and MLCommons’ social media channels, but we’ve learned that building an inclusive community requires going further. In 2024, we expanded our reach by leveraging the mailing list from the undergraduate architecture mentoring workshop (uArch), focusing on institutions with graduate programs. Also, we actively partner with affinity groups’ social media networks, including the Computer Architecture Student Association (CASA), Women in Architecture (WicArch), Indigenous in AI, Black in AI, LatinX in AI, Women in ML, and CRA-WP.
Two-Day Program
The main part of our program is a two-day workshop that combines structured learning with organic networking opportunities. Rather than following a traditional conference format, we create an environment where Rising Stars can engage with their peers and senior researchers. The feedback has been positive:
“I loved the connections I made,” reflects a 2024 participant.
One particularly successful aspect of the workshop has been our “Ask Me Anything” sessions with distinguished researchers. These candid discussions provide insights into career pathways and research directions that aren’t typically available in more formal settings. For instance, in previous workshops, participants engaged in frank discussions about career trajectories with luminaries in our field, like Turing Award winner Prof. David Patterson and Prof. Margaret Martonosi, gaining valuable perspectives on navigating academic and industry careers. Rising Stars also got to meet with
“The speakers and the speaking session were very good, well thought of and well planned. I learned a lot from it. It was also a great idea to add the poster session. I had a great time presenting my research and receiving feedback from everyone.”
The program recognizes the importance of industry engagement in ML systems research. Students gain insights into real-world deployment challenges and research priorities with the assistance of MLCommons industry members, who are invited to participate in activities with the students at no cost. These connections have fostered meaningful collaborations, leading several participants to secure research internships and project partnerships through the network they built at the workshop.
Impact and Future Directions
Each year, we gather feedback from the Rising Stars cohort, workshop attendees, and collaborating institutions on continuing to strengthen and grow the Rising Stars program. This feedback has proved invaluable in tailoring the workshop and experience to emerging researchers in ML and computer systems. The impact of the ML & Systems Rising Stars program extends beyond the workshop itself.
“The most exciting thing was, I got to know more about the most cutting-edge research in other domains, especially in efficient ML,” shares a 2024 participant.
These cross-domain connections have led to concrete outcomes: “Connections with other rising stars expand my AI research from edge to cloud,” reports a 2023 participant, “and one connection brings my student at UCF an internship opportunity this summer.”
“I found vast opportunities to investigate security and privacy issues within the domain.”
Feedback from our 2024 cohort has also been particularly illuminating. Participants consistently highlighted the value of interdisciplinary exposure, noting how discussions with peers working on different aspects of ML systems helped them identify new research directions. As one Rising Star noted:
“The workshop helped me understand cutting-edge developments across the field and highlighted the need for practical applications of research. The diverse perspectives from both academia and industry provided a comprehensive understanding of current challenges and future directions.”
The program’s industry partnerships have proved incredibly valuable in unexpected ways. Beyond providing venues for hosting the workshop—hosted by Google in 2023, NVIDIA in 2024, and upcoming at Meta in 2025—these partnerships have created opportunities for collaboration between academia and industry.
Call to Action
Looking ahead to 2025, we are excited to announce several enhancements to the program. First, we are partnering with the MLSys conference, scheduling our workshop for May 7-8, immediately preceding the main conference. This timing will allow the 2025 Rising Stars to engage more deeply with the broader ML systems research community. Second, we are expanding our mentorship component to provide more sustained support throughout the year, helping participants navigate both technical and career challenges.
As we prepare for the 2025 workshop, we invite the SIGARCH community to engage with this initiative:
- For potential applicants: Applications for the 2025 cohort opened on December 16, 2024 (see below), and close on January 31, 2025. We particularly encourage applications from researchers working to bridge different aspects of ML systems, whether in academia, industry, or national laboratories.
- For faculty members: Consider nominating your promising students and postdocs for the program. Your recommendation letters provide context that helps the selection committee identify candidates who can both contribute to and benefit from the community of ML Systems engineers we’re building.
- For industry researchers: We welcome your participation as mentors and technical advisors. Your experience deploying ML systems into the real-world provides valuable perspectives for young researchers.
The program details and application materials are available at https://mlcommons.org/about-us/programs/.
Conclusion
As we work to formalize and expand the ML & Systems Rising Stars program, we hope to gather feedback from the community and increase the awareness of this effort. We appreciate the industry support we have received thus far and look forward to additional future collaborations at the uniquely cross-industry-academic-research efforts in ML & Systems. In future years, we hope to maintain and expand our collaboration with MLCommons and industry partners, track the trajectory and collaborations that spring from our incredible cohorts, and extend similar mentoring and community-building opportunities to different stages of researchers, such as undergraduate and early-PhD students. If you have thoughts on improving this program or would like to contribute, please get in touch with the organizing committee!
Acknowledgments
Thanks to our incredible volunteers’ generous dedication, time, and effort. In addition to our supporting organizations, we’d like to thank everyone who has served on the program committee: Newsha Ardalani (Meta), Ruichuan Chen (Nokia Bell Labs), Qirong Ho (MBZUAI), Jenny Huang (NVIDIA), Tianyu Jia (Peking University), Zhihao Jia (Carnegie Mellon University), Sangeetha Abdu Jyothi (University of California Irvine), Salman Khan (MBZUAI), Ana Kilmovic (ETH Zurich), Tushar Krishna (Georgia Institute of Technology), Nils Lukas (MBZUAI), Kanak Mahadik (Adobe), Josh San Migual (University of Wisconsin-Madison), Akshay Nambi (Microsoft Research), Gennady Pekhimenko (University of Toronto), Lillian Pentecost (Amherst College), Gilles Pokam (Intel), Chris Re (Stanford University), Brandon Reagen (New York University), Mark Ren (NVIDIA), Sasha Rush (Cornell Tech/Hugging Face), Hashim Sharif (AMD), Virginia Smith (Carnegie Mellon University), David Stutz (IBM Research), Thierry Tambe (Stanford University), Devashree Tripathi (IIT Bhubaneswar), Shivaram Venkataraman (University of Wisconsin-Madison), Emma Wang (Google), Yu Wang (Tsinghua University), Francis Yan (University of Illinois Urbana-Champaign), Amir Yazdanbakhsh (Google), Eiko Yoneki (University of Cambridge), Jeff Zhang (Arizona State University), Jiawei Zhao (Meta FAIR). We would also like to thank individuals who helped organize the Rising Stars program, including Kelly Berschauer (MLCommons), David Kanter (MLCommons), Rebecca Weiss (MLCommons), Heather Brojer (Google), Naveen Kumar (Google), Peter Mattson (Google), Ritika Borkar (NVIDIA), Petrina Mitchell (Meta), and Vikas Chandra (Meta).
About the Authors
Akanksha Atrey is a Research Scientist at Nokia Bell Labs. Her work lies at the intersection of artificial intelligence and distributed systems, focusing on building privacy-preserving, trustworthy, and resource-efficient edge AI and web3 technologies.
Sercan Aygun is an Assistant Professor at the University of Louisiana at Lafayette. He specializes in tiny machine learning and emerging computing paradigms, including stochastic and hyperdimensional computing, for the lightweight design of learning systems.
Udit Gupta is an Assistant Professor at Cornell Tech. His research interests lie at the intersection of computer architecture, systems, machine learning and environmental sustainability.
Abdulrahman Mahmoud is an Assistant Professor at MBZUAI in Abu Dhabi His work is at the intersection of computer architecture, software system design, and machine learning, with the goal of co-designing future ML systems for high performance, scalable reliability, and intelligent resource allocation.
Lillian Pentecost is an Assistant Professor at Amherst College. Her research aims to improve memory system efficiency by integrating emerging technologies and developing new design methods and tools.
Vijay Janapa Reddi is a Professor at Harvard University and co-founder/Vice President of MLCommons. He is passionate about research and education in computer architecture, machine learning systems, and autonomous agents. To help make ML systems education more accessible, consider giving MLSysBook.AI a GitHub ⭐. Each star not only supports learners but also raises donations for education and outreach in developing countries.
Disclaimer: These posts are written by individual contributors to share their thoughts on the Computer Architecture Today blog for the benefit of the community. Any views or opinions represented in this blog are personal, belong solely to the blog author and do not represent those of ACM SIGARCH or its parent organization, ACM.
