Pratyush Das

das160@purdue.edu

I am a PhD student in Computer Science at Purdue University studying compilers, advised by Professor Milind Kulkarni and Professor Ben Delaware.

My research interests lie in using staging compilers to improve the performance of high performance systems. I have looked at improving the performance of operations over files stored in distributed file systems (NSDI'23, GPCE'24) and I am currently looking into further extending the idea to speed up sparse tensor computations (arxiv).

During the summer of 2023, I interned at Apple, working on a custom target and runtime for the Swift programming language. As an undergrad, I interned with Princeton University for multiple years, working on software for High Energy Physics. During this period, I was a core developer of two software projects that found significant adoption in the Physics community - Awkward Array (library for irregularly shaped tensor computations) and uproot (Python re-implementation of I/O for physics motivated by interoperability with other existing Python packages). I also made significant contributions to the following LLVM based projects: Enzyme, Cling and Clang.

My CV can be found here.

Send me an email if you want to give a talk at PurPL (Purdue University's Programming Languages Group).

News

  • 10-28-2024: Presenting poster about SABLE at MWPLS 2024.
  • 08-29-2024: Took over as a student co-ordinator of the Purdue Programming Languages group.
  • 08-15-2024: Our paper on specializing distributed storage accesses was accepted to GPCE 2024.
  • 12-24-2023: I will be interning at Apple this summer.
  • 12-09-2022: Our paper DiSh was accepted to NSDI 2023.
  • Publications

    Talks


    Teaching

    I have been a TA for the following courses at Purdue University: