Niranjana Deshpande

prof_pic.jpg

I am a data and applied scientist at Microsoft, where I analyze M365 Copilot usage patterns to drive product growth. I received my Ph.D. from Rochester Institute of Technology.

My dissertation research – Towards Algorithm Selection for Efficient Search-Based Software Engineering – centered on leveraging algorithm selection for software engineering tasks. The goal of my research was to enable practitioners in building and maintaining software systems more effectively, using specific optimization algorithms for different problem instances. My work combined empirical analysis of evolutionary algorithms on hard SE problems, training ML models to predict algorithm performance, and using those models to build computationally efficient and functionally correct software systems.

I’ve applied these techniques to service-oriented systems and automated API recommendations during library migration. Our paper on self-adaptive service-oriented systems using algorithm selection won the best paper award at IEEE ICWS 2021.​

news

Jul 15, 2024 I’ve joined Microsoft as a Data and Applied Scientist, in Redmond.
Feb 02, 2024 Website revamp successful.
Nov 27, 2023 Our paper on evaluating multi-objective evolutionary algorithms for API migration was accepted to the Journal of Swarm and Evolutionary Computation.
Oct 04, 2023 I’m presenting my research at the Doctoral Symposium at ESEC/FSE in San Francisco this December.
Apr 15, 2023 I’ve accepted a data scientist internship in the Product Analytics and Data Science (PANDAS) team at Amazon Web Services in Seattle.
May 14, 2022 I’ve successfully defended my thesis proposal, and am looking forward to my internship in Seattle!

selected publications

  1. ICWS
    R-CASS: Using algorithm selection for self-adaptive service-oriented systems
    Niranjana Deshpande ,  Naveen Sharma ,  Qi Yu , and 1 more author
    In 2021 IEEE International Conference on Web Services (ICWS) , 2021
  2. EvoApps
    Search-based third-party library migration at the method-level
    Niranjana Deshpande ,  Mohamed Wiem Mkaouer ,  Ali Ouni , and 1 more author
    In International Conference on the Applications of Evolutionary Computation (Part of EvoStar) , 2022
  3. ICWS
    Online Learning Using Incomplete Execution Data for Self-Adaptive Service-Oriented Systems
    Niranjana Deshpande ,  Naveen Sharma ,  Qi Yu , and 1 more author
    In 2022 IEEE International Conference on Web Services (ICWS) , 2022
  4. SEC
    Third-party software library migration at the method-level using multi-objective evolutionary search
    Niranjana Deshpande ,  Mohamed Wiem Mkaouer ,  Ali Ouni , and 1 more author
    Swarm and Evolutionary Computation, 2024