Hi, I am Prasun and I am involved with different open-source projects aimed at super fast Scientific Computing on Ruby and D. I am a huge supporter of Open-source software and love Bioinformatics. I am highly motivated to bring Scientific Computing to Ruby.

I work as a Software Architect at Modak Analytics Software. Here, I am responsible for building Genome scan tools optimized for GPU clusters, targetted for Big Pharma.

I am an active member of the Genenetwork team with University of Tennessee Health Science Center, building Faster-LMM-D, a GWAS tool for HPC devices. Faster_LMM_D supports Linear Mixed Models. Bayesian Sparse Models are on the way. Faster_LMM_D has a GPU backend that helps it run in CUDA and OpenCL environment.

I am a Ruby Grant 2017 recipient. Ruby Grant is awarded every year to 3-5 projects by Ruby Association, Japan. It helped me build RbCUDA: CUDA bindings for Ruby. RbCUDA is the fastest Ruby library for matrix multiplication.

I worked as a student developer for Ruby Science Foundation for Google Summer of Code 2016 on the project “Port NMatrix to JRuby”.

I worked as a student developer for Ruby Science Foundation for Google Summer of Code 2017 on the project “Creating the fastest math libraries for Ruby/JRuby by using the GPU through OpenCL and ArrayFire”.

Educational Qualifications

  • M.Sc., Biological Sciences (2012 - 2017)
    Birla Institute of Technology and Science, Pilani, India

  • B.E., Chemical Engineering (2012 - 2017)
    Birla Institute of Technology and Science, Pilani, India

Programming

I like to program in C, C++, Java, D and Ruby.

I love GPU programming and have built opensource libraries in D and Ruby that help in GPU acceleration. I am the creator and maintainer of ArrayFire-rb, ArrayFire-D, cuda_d and RbCUDA.

I am highly-skilled in building data-intensive applications and speeding up the algorithms by optimizing the underlying code, introducing compiler level optimizations, code profiling, exploiting multicore CPUs and GPUs.

Very recently, I have developed an interest in Deep Learning and AI.

Talks

  1. Invited as a Speaker at RubyKaigi 2018, Sendai, Japan to talk about High Performance GPU Computing with Ruby.

  2. Speaker at Ruby Conference India, 2018 to talk about GPU Accelerated Libraries for Ruby.
    Slides| Video

  3. Speaker at RubyConf 2017, New Orleans, USA to talk about High Performance GPU Computing on Ruby.
    Slides| Video

  4. Speaker at RubyWorld Conference, Matsue Japan, 2017 to talk about High Performance GPU Computing on Ruby.
    Slides| Video

  5. Speaker at Ruby DevRoom at FOSDEM 2017, Brussels, Belgium to talk about Scientific Computing on JRuby.
    Slides| Video

  6. Speaker at Ruby Conference India, 2017 to talk about Scientific Computing on JRuby.
    Slides| Video

Posters

  1. Poster on Faster Linear Mixed Models (LMM) for online GWAS omics analysis​ at Complex Trait Community Conference 2017, Memphis, Tennessee, USA.
    Poster

  2. Poster on RbCUDA: CUDA Bindings for Ruby at Ruby 25th Event, Tokyo, Japan.
    Poster

Awards

  1. Finalist for Fukuoka Ruby Award 2018 for activity: Ruby for High Performance Scientific Computing using Arrayfire gem and RbCUDA gem.
  2. Ruby Association Travel Grant 2017

Achievements

  1. Ruby Grant 2017
  2. Google Summer of Code 2017
  3. Google Summer of Code 2016

Mentoring

  1. Google Summer of Code 2018: Mentoring two students on the development of “Ruby port of Matplotlib”, with Ruby Science Foundation.

Projects

  1. RbCUDA
  2. ArrayFire-rb
  3. NMatrix
  4. ArrayFire-d
  5. Faster_lmm_d
  6. gperftools_d
  7. Dmatrix
  8. Mitab parser

Other interests

Swimming, Gaming, Travelling