Haimasree Bhattacharya, Ph.D.

Senior Research Engineer at The Hebrew University of Jerusalem
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
Israel, IL
Languages
  • English Full professional proficiency
  • Bengali Native or bilingual proficiency
  • Hindi Limited working proficiency

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
You need to have a working account to view this content.

Credentials

  • PostgreSQL for Everybody
    Coursera
    Nov, 2021
    - Oct, 2024
  • Machine Learning Engineering for Production (MLOps)
    Coursera
    Sep, 2021
    - Oct, 2024
  • Practical Data Science by Amazon Web Services & DeepLearning.AI on Coursera
    Coursera
    Sep, 2021
    - Oct, 2024
  • Deep Learning Specialization
    Coursera
    Jun, 2021
    - Oct, 2024

Experience

    • Israel
    • Higher Education
    • 700 & Above Employee
    • Senior Research Engineer
      • Oct 2022 - Present

    • Germany
    • Research Services
    • 700 & Above Employee
    • Scientific Programmer
      • Aug 2020 - Jul 2022

      I primarily worked on research, development and maintenance of Kipoi - a model zoo for predictive sequence based models in genomics and a unified API to access these models. Some of the projects that I worked on are as follows - kipoi-veff2 - A python package to perform Variant Effect Prediction (VEP) with a subset of models in Kipoi. This package provides a simple, easy to use and extendable API to perform inference and scoring with ∼1500 Kipoi models trained with a variety of deep learning frameworks such as Tensorflow, Pytorch and Keras. Furthermore, Snakemake based generic workflows are also provided to help run VEP at scale in hpc clusters. kipoi-containers - A python package to containerise Kipoi model environments into docker and singu-larity images and dissemination of these containerised images through Dockerhub and Zenodo following FAIR principles. Additionally, an infrastructure to support automated update and addition of docker and singularity images as models get updated/added in Kipoi has also been developed. Development and maintenance of Kipoi and model zoo: • Feature development and new releases of Kipoi python package. Some of the added features are- support for Tensorflow 2.* and Keras 2.* models in Kipoi, adding a cli option to run a Kipoi model directly within its singularity images, keeping up with increasing python versions, ability to output metadata along with predictions from inference on user data etc. • CI monitoring, failure resolution and maintenance across several repositories in Kipoi. • Addressing user questions and comments and facilitating addition of new model groups in Kipoi. Show less

    • Germany
    • Research Services
    • 700 & Above Employee
    • Scientific Staff
      • Aug 2020 - Jul 2022

    • Netherlands
    • Hospitals and Health Care
    • 700 & Above Employee
    • Research Software Engineer
      • Apr 2018 - Jul 2020

      I was a member of research and software development team in Diagnostic Image Analysis Group, RadboudUMC. Some of the projects that I worked on are as follows. My primary work was in the area of automatic detection and characterization of lung nodules in chest CT. This project is partially funded by and carried out in close collaboration with MeVis Medical Solutions (MMS). This project resulted in the release of Veolity, an efficient workstation for reading lung CT scans. I incorporated BTS and Fleischner guidelines for pulmonary nodule screening and management into the workstation. I was also responsible for the development and maintenance of CIRRUS Lung Screening, research prototype of Veolity. I contributed to Grand Challenge, a platform to objectively evaluate biomedical imaging algorithms. In particular, I developed several REST APIs in grand challenge that allows users to upload images, run algorithms on them and download the results via a python client. I also worked on a workstation based on MeVisLab that allowed users to view and interact with medical images on grand challenge. I translated the research code that secured a runner up solution of Kaggle Data Science Bowl 2017 (https://dhammack.github.io/kaggle-ndsb2017/) into an automated easy to use deep learning pipeline for in house deep learning cluster. I further augmented this pipeline to handle compressed DICOM and ITK file formats. Publication: https://pubs.rsna.org/doi/10.1148/ryai.2021210027 Show less

    • Research
    • 700 & Above Employee
    • Scientific co-worker
      • Dec 2016 - Apr 2018

      Developed an interactive web-based python tool for performing differential gene expression using Allen Brain API and JUBrain Atlas which has been deployed in the Neuroinformatics platform of the Human Brain Project. Developed an interactive web-based python tool for performing differential gene expression using Allen Brain API and JUBrain Atlas which has been deployed in the Neuroinformatics platform of the Human Brain Project.

    • South Korea
    • Computers and Electronics Manufacturing
    • 700 & Above Employee
    • Technical Lead
      • Oct 2014 - Feb 2016

      Led a team of 3 people in the multimedia group, Advanced Technology Lab (ATL) on various computer vision and image processing research projects for their flagship phone Galaxy. Led a team of 3 people in the multimedia group, Advanced Technology Lab (ATL) on various computer vision and image processing research projects for their flagship phone Galaxy.

    • United States
    • Higher Education
    • 700 & Above Employee
    • Graduate Research Assistant
      • Jun 2009 - Aug 2014

      Conducted research on the use of physically based modeling techniques for computer animation.Dan Gerszewski, Haimasree Bhattacharya, and Adam W. Bargteil. A Point-based Methods for Animating Elastoplastic Solids." In the proceedings of ACM SIGGRAPH/Eurographics Symposium on Computer Animation. New Orleans, Louisiana, September 2009.Haimasree Bhattacharya, Yue Gao and Adam W. Bargteil. A Level-set Method for Extracting Surfaces from Animated Particle Data" In the proceedings of ACM SIGGRAPH/Eurographics Symposium on Computer Animation 2011Haimasree Bhattacharya, Michael Nielsen, Robert Bridson,Steady State Stokes flow Interpolation for Fluid Control. Eurographics 2012, short papersHaimasree Bhattacharya, Joshua A Levine, Adam W. Bargteil,Fluid Simulation on Unstructured Quadrilateral Surface Meshes.Accepted to the workshop "Structured Meshing: Theory, Applications, and Evaluation"Haimasree Bhattacharya, Yue Gao, Adam W. Bargteil,A Level-set Method for Skinning Animated Particle Data.Accepted to IEEE Transactions on Visualization and Computer GraphicsHaimasree BhattacharyaPhysics-Based Animation Tools For Fluids And Surfaces: Generating Surfaces From Fluids, Animating Fluids On Surfaces And Controlling Fluids From Surfaces.PhD Dissertation Show less

    • Teaching Assistant
      • Aug 2012 - Dec 2012

    • Teaching Assistant
      • Jan 2012 - May 2012

    • New Zealand
    • Entertainment Providers
    • 300 - 400 Employee
    • Simulation RnD Intern
      • Jun 2011 - Sep 2011

      Worked in the Simulation research and development group developing a Naiad plugin for a Fluid control Problem based on Stokes Flow. Haimasree Bhattacharya, Michael B. Nielsen and Robert Bridson "Steady State Stokes Flow Interpolation for Fluid Control" to appear in Eurographics 2012, short papers. Worked in the Simulation research and development group developing a Naiad plugin for a Fluid control Problem based on Stokes Flow. Haimasree Bhattacharya, Michael B. Nielsen and Robert Bridson "Steady State Stokes Flow Interpolation for Fluid Control" to appear in Eurographics 2012, short papers.

    • United States
    • Animation and Post-production
    • 700 & Above Employee
    • FX RnD Intern
      • Jun 2010 - Sep 2010

      Worked in the FX research and development group developing a tool for seamlessly deforming shape of an input piece of cloth, so that it matches given artist drawn 2d silhouette. Worked in the FX research and development group developing a tool for seamlessly deforming shape of an input piece of cloth, so that it matches given artist drawn 2d silhouette.

    • United States
    • Higher Education
    • 700 & Above Employee
    • Teaching Assistant
      • Aug 2008 - May 2009

Education

  • University of Utah
    PhD, Computer Science
    2008 - 2014
  • Jadavpur University
    BE, Computer Science
    2004 - 2008

Community

You need to have a working account to view this content. Click here to join now