Background

• Employment:

• Postdoctoral Research Associate at the University of Cambridge, United Kingdom.
      Computer Architecture group
      Advisor: Prof. Dr. Timothy Jones

• Education:

  • PhD (2018 - Jan, 2024) - University of Delaware, USA
    Advisor: Prof. Dr. Rudolf Eigenmann
  • B.E (2012 - 2016) - University of Mumbai, India

Formal CV

You can find my CV here : Akshay’s CV

Principal Research Interests

• Compiler Optimization techinuqes
• Hardware-Software co-design
• Automatic Parallelization
• High Performance Computing

Publications

Conference Proceedings

• Márton Erdős, Utpal Bora, Akshay Bhosale, Alexandra W Chadwick, Bob Lytton, Giacomo Gabrielli, Richard Cooper, Yuxin Guo, and Timothy M. Jones, “Loopfrog: In-core hint-based loop parallelization,” in Proceedings of The 58th IEEE/ACM International Symposium on Microarchitecture, Seoul, Korea, 2025, (Accepted to appear).

• Yuxin Guo, Akshay Bhosale, Alexandra W Chadwick, Utpal Bora, Márton Erdős, Giacomo Gabrielli, and Timothy M. Jones, “Ghost threading: Helper-thread prefetching for real systems,” in Proceedings of The 58th IEEE/ACM International Symposium on Microarchitecture, Seoul, Korea, 2025, (Accepted to appear).

• Akshay Bhosale and Rudolf Eigenmann. Recurrence analysis for automatic parallelization of subscripted subscripts in Proceedings of the 29th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, ser. PPoPP ’24, Edinburgh, United Kingdom: Association for Computing Machinery, 2024, pp. 392–403, isbn: 97984007043522403.

• Akshay Bhosale and Rudolf Eigenmann. On the automatic parallelization of subscripted subscript patterns using array property analysis. In Proceedings of the ACM International Conference on Supercomputing (ICS ‘21). Association for Computing Machinery, New York, NY, USA, 392–403. 2021. Erratum

Journal Articles

• Akshay Bhosale, Parinaz Barakhshan, Miguel Romero Rosas. Automatic and Interactive Program Parallelization Using the Cetus Source to Source Compiler Infrastructure v2.0. Special Issue “Program Analysis and Optimizing Compilers for High-Performance Computing”, Electronics. 2022; 11(5):809.

Workshop proceedings

• Akshay Bhosale and Rudolf Eigenmann, “Compass: A combined parallel subscripted subscript benchmark suite,” in 36th International Workshop on Languages and Compilers for Parallel Computing (LCPC), Lexington, KY, USA, October 11–13, (Accepted to appear), 2023, p. 221.

• Akshay Bhosale, and Rudolf Eigenmann. Compile-time Parallelization of Subscripted Subscript Patterns 2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW). IEEE, 2020.

Patents

• Richard Cooper, Giacomo Gabrielli, Bob Lytton, Marton Erdos, Alexandra Winifred Chadwick, Akshay Bhosale, Utpal Bora, and Timothy Jones, “Handling reductions in micro-threaded code,” Patent Application No.: 202 511 074 400, Application filed in India, 2025.

• Giacomo Gabrielli, Bob Lytton, Richard Cooper, Alexandra Winifred Chadwick, Marton Erdos, Yuxin Guo, Utpal Bora, Akshay Bhosale, and Timothy Jones, “Support for parallel function continuations,” Patent Application No.: 202 511 074 398, Application filed in India, 2025.

• Giacomo Gabrielli, Bob Lytton, Ali Zaidi, Utpal Bora, Akshay Bhosale, Marton Erdos, and Timothy Jones, “Memory aliasing discriminators,” Patent Application No.: 202 511 074 399, Application filed in India, 2025.

Published Posters

• Alexandra W. Chadwick, Márton Erdős, Utpal Bora, Akshay Bhosale, Bob Lytton, Yuxin Guo, Richard Cooper, Giacomo Gabrielli, and Timothy M. Jones, The future of instruction-level parallelism (ilp), 2025 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2025.

Projects

Speculative Task-Parallelization

The ParaSol project funded by the Engineering and Physical Sciences Research Council (EPSRC) UK, seeks to develop compiler analyses and transformations to extract parallelism for future processors. I successfully implemented optimization techniques (in C++) to take advantage of and improve instruction-level parallelism for efficient speculative execution of application codes. The techniques have been incorporated into the ParaSol compiler, developed using the industry standard LLVM infrastructure.

Software Prefetching

Implemented a compiler transformation pass (in LLVM) to automatically extract helper thread for prefetching data in a Simultaneous Multi-threading (SMT) execution context.

Advanced Compiler Analysis Techniques for Irregular Computations

Implemented an Analysis technique and extended the related Dependence Test for compile-time automatic parallelization of Subscripted Subscript patterns using the Cetus Source to Source translator which translates from C to C annoted with OpenMP. Details in the publications listed above.

Professional Service

• International Symposium on Code Generation and Optimization (CGO) - ACM Student Research Competition (SRC), 2025, Las Vegas, USA.

• The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC), 2024, Atlanta, USA.

Supervision Experience

• Course Supervisor, Lent Term 2025, Course on Optimizing Compilers, Department of Computer Science and Technology, University of Cambridge.

• Supervisor, Undergraduate Final Year Project titled “A JIT Compiler for BEAM bytecode to RISC-V”

Internships

• At the Pacific Northwest National Lab
  In this project we implemented a Python frontend for a Domain Specific Compiler base on MLIR for Computational Chemistry applications. The compiler is called COMET.

Leadership

• University of Delaware
  • VP Student Affairs (2020 - 2021) - Graduate Student Government
  • International Student Leader (2018 - 2019) - Office of International Students and Scholars
  • Team Lead - Blue Hen iBuddy Global Mentoring Program

Contact Info

Department of Computer Science
William Gates Building, JJ Thompson Ave
Room SC07
University of Cambridge
Cambridge,Cambridgeshire,UK

   asb227@cam.ac.uk

    Linkedin

    Github