AI People Wiki

Builders & open source

Priya Nagpurkar

Vice President of AI Platform (as titled by IBM Research in March 2026), IBM

  • Vice President of AI Platform (as titled by IBM Research in March 2026), IBM Research1

Priya Nagpurkar is a vice president at IBM Research. IBM Research's March 2026 post on contributing the llm-d distributed inference framework to the Cloud Native Computing Foundation quotes Nagpurkar as vice president of AI platform at IBM Research. In 2023, IBM Research described Nagpurkar as vice president of hybrid cloud platform and developer productivity, overseeing the research behind IBM's hybrid cloud platform. Nagpurkar is a co-author, listed with IBM Research, of a 2025 paper on category-aware semantic caching for LLM serving and of a 2024 IBM Research and University of Illinois white paper on transforming hybrid cloud systems for AI workloads.12345

At a glance

Current role
Vice President of AI Platform (as titled by IBM Research in March 2026)
Company
IBM
Sector
Company
Work listed
2
Sources cited
5
Profile reviewed

Career

In 2023 IBM Research described Nagpurkar as vice president of hybrid cloud platform and developer productivity, overseeing the research behind IBM's hybrid cloud platform; that year Nagpurkar spoke for IBM when it joined the PyTorch Foundation. By March 2026 IBM Research titled Nagpurkar vice president of AI platform, in its announcement that IBM Research, Red Hat, and Google Cloud were contributing the llm-d distributed inference framework to the Cloud Native Computing Foundation.231

Research

Nagpurkar's recent papers concern systems for AI workloads. A 2024 white paper by IBM Research and University of Illinois Urbana-Champaign researchers sets out a full-stack co-design approach to hybrid cloud systems for AI, and a 2025 paper on category-aware semantic caching proposes LLM-serving cache policies whose similarity thresholds, time-to-live, and quotas vary by query category.54

Work and contributions

  • Category-aware semantic caching for LLM workloads Published research

    Co-author of "Category-Aware Semantic Caching for Heterogeneous LLM Workloads" (2025), which proposes caching policies that vary similarity thresholds, time-to-live, and quotas by query category in LLM serving systems.4

  • Transforming the Hybrid Cloud for Emerging AI Workloads Published research

    Listed with IBM Research among the authors of a 2024 white paper, developed by IBM Research and University of Illinois Urbana-Champaign researchers, on full-stack co-design of hybrid cloud systems for AI workloads.5

Timeline

  1. Represented IBM Research as vice president of hybrid cloud platform when IBM joined the PyTorch Foundation.2
  2. Co-authored "Transforming the Hybrid Cloud for Emerging AI Workloads" with IBM and Illinois researchers.5
  3. Co-authored "Category-Aware Semantic Caching for Heterogeneous LLM Workloads".4
  4. Quoted as IBM Research vice president of AI platform as llm-d was contributed to the CNCF.1

Other documented roles

Roles the sources document that are not confirmed as current.

  • Vice President of Hybrid Cloud Platform and Developer Productivity, IBM Research23

Where they work

  • IBM 29 people in the wiki

Suggest a correction or removal

Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.

References

5 sources · profile reviewed Oct 21, 2026 · data updated Oct 21, 2026

  1. Donating llm-d to the Cloud Native Computing Foundation (IBM Research blog)IBM Research · Announcement · published Mar 24, 2026 · accessed Oct 21, 2026
  2. IBM joins the PyTorch Foundation to further open-source AI collaboration (IBM Research blog)IBM Research · Announcement · published Jul 27, 2023 · accessed Oct 21, 2026
  3. How open source paved the way for computing from anywhere (IBM Research blog)IBM Research · Official page · published Sep 20, 2023 · accessed Oct 21, 2026
  4. Category-Aware Semantic Caching for Heterogeneous LLM Workloads (arXiv 2510.26835)arXiv · Paper · published Oct 29, 2025 · accessed Oct 21, 2026
  5. Transforming the Hybrid Cloud for Emerging AI Workloads (arXiv 2411.13239)arXiv · Paper · published Nov 20, 2024 · accessed Oct 21, 2026