Jovan Stojkovic

Jovan Stojkovic is an incoming Assistant Professor at the Department of Computer Science at the University of Texas at Austin, set to join in Fall 2026. Prior to his appointment at UT Austin, he will spend a year at Meta working with the AI and Systems Co-design group. His research focuses on cloud computing and datacenters, with particular emphasis on cloud-native workloads and machine learning inference.

Education:

• PhD in Computer Science from the University of Illinois at Urbana-Champaign, advised by Professor Josep Torrellas
• Undergraduate studies at the School of Electrical Engineering, University of Belgrade, Serbia, where he was recognized as the best student of the Computer Engineering and Information Theory Department every year from 2017-2020

Research Interests:

Jovan's research focuses on cloud computing and datacenters, with two primary domains:

• Cloud-native workloads, such as microservices and serverless computing. He investigates how to co-design novel hardware platforms and software systems that deliver orders-of-magnitude improvements in performance, energy efficiency, and resource utilization for these emerging workloads.
• Machine Learning (ML) inference, particularly large language models (LLMs). His work addresses the challenges of ML inference through smart scheduling, workload placement, and system-level configuration tuning to reduce energy, power, and thermal overheads while maintaining performance and accuracy guarantees.

Publication Trends:

Jovan's publications demonstrate a strong focus on optimizing cloud infrastructure for emerging workloads. His research spans across serverless computing, microservices, and large language model inference. A clear trend emerges in his work: addressing the performance, energy efficiency, and resource utilization challenges of modern cloud workloads through innovative hardware-software co-design approaches. His most recent work shows increasing focus on LLM inference optimization, particularly in the areas of thermal management, power efficiency, and scheduling for many-adapter environments.

Awards and Honors:

• HPCA Best Paper Award (2025)
• IEEE MICRO Top Picks Honorable Mention (2024)
• 6 patents with IBM and Microsoft on: Serverless systems, Processor overclocking in the cloud, and Energy-efficient LLM inference
• W. J. Poppelbaum Memorial Award (2025) for hardware and architecture innovation
• Mavis Future Faculty Fellowship (2024–2025)
• Invited to present at 11th Heidelberg Laureate Forum (2024)
• Kenichi Miura Award (2022) for excellence in High Performance Computing
• Multiple student travel grants to ISCA, MICRO, ASPLOS, and HPCA

Advising and Grants:

Jovan is actively seeking prospective PhD students for his research group at UT Austin. His research has been supported through collaborations with major tech companies including IBM, Microsoft, and Meta. His six patents with IBM and Microsoft demonstrate the practical impact of his research in serverless systems, processor overclocking, and energy-efficient LLM inference. His work on serverless computing (MXFaaS, EcoFaaS) and LLM inference optimization has received significant recognition in top-tier computer architecture conferences.

Research Groups:

During his PhD at UIUC, Jovan worked with Professor Josep Torrellas on cloud infrastructure research. He has collaborated extensively with researchers at IBM Research (particularly Hubertus Franke) and Microsoft (particularly Íñigo Goiri and Ricardo Bianchini). His upcoming position at UT Austin will establish his independent research group focused on cloud computing and datacenter systems. His year at Meta working with the AI and Systems Co-design group will further strengthen his expertise in AI infrastructure.