Saumik Dana, Ph.D.

Computational Engineer at Advanced Scanners
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
Austin, Texas, United States, US

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.

Experience

    • United States
    • Medical Equipment Manufacturing
    • 1 - 100 Employee
    • Computational Engineer
      • Aug 2023 - Present

      → Python code to interface with Kuka robot → Python code to interface with Kuka robot

    • United States
    • Software Development
    • 1 - 100 Employee
    • Computational Lead
      • Aug 2022 - Mar 2023

      → Developed a Python-based backend for a stock price prediction framework → Engineered a forecasting model through a process involving the discovery of a PDE for each stock price time series using a sparse regression greedy algorithm, XGBoost and Bayesian parameter optimization → Used autoregression analysis to establish the optimal window size for training the model → Exhibited exceptional generalization through comprehensive cross validation on the Kaggle stock market dataset → Collaborated with a ML engineer to conduct extensive unit testing and continuous delivery integration using GitHub Actions Show less

    • United States
    • Higher Education
    • 700 & Above Employee
    • Postdoctoral Research Associate
      • Nov 2020 - Jul 2022

      → Implemented a two-grid algorithm in a Python/C++ codebase with SWIG bindings, optimizing the compute time for large-scale grid problems by 50% → Utilized CMake to automate the build systems and conducted simulations on AWS EC2 instances → Created separate Python and C++ based codes for model parameter estimation from time series data using Bayesian inference → Diagnozed the performance of both MySQL and JSON in handling time series data interchange between the generation and inference modules → Utilized Markov chain Monte Carlo (MCMC) sampling using the Metropolis Hastings (MH) algorithm to infer model parameter distribution Show less

    • United States
    • Higher Education
    • 700 & Above Employee
    • Postdoctoral Research Associate
      • Feb 2020 - Oct 2020

      → Implemented algorithms for short and long read sequencing using Pythonic string manipulation libraries → Collaborated with biologists to develop user-friendly code for a forensics project → Implemented algorithms for short and long read sequencing using Pythonic string manipulation libraries → Collaborated with biologists to develop user-friendly code for a forensics project

    • United States
    • Higher Education
    • 700 & Above Employee
    • Postdoctoral Research Associate
      • Aug 2019 - Jan 2020

      → Generated simulations of fluid flow to provide recommendations to a NYC based startup on design of a vertical axis wind turbine → Demonstrated a 20% power loss reduction through design improvements, focusing on the fillet region of the vertical axis wind turbine → Generated simulations of fluid flow to provide recommendations to a NYC based startup on design of a vertical axis wind turbine → Demonstrated a 20% power loss reduction through design improvements, focusing on the fillet region of the vertical axis wind turbine

    • United States
    • Research Services
    • 700 & Above Employee
    • Postdoctoral Research Associate
      • Jan 2019 - Jul 2019

      → Collaborated with scientists to integrate a Python wrapper that implements a graph theory based reduced order model for subsurface flow and transport → Traded 3D geometry for a graph-based reduced system that was 500-1000 times faster in terms of computational efficiency for this specific application → Showed that the reduced-order model has great potential for providing operators with real-time decision-making information for optimizing hydrocarbon production → Collaborated with scientists to integrate a Python wrapper that implements a graph theory based reduced order model for subsurface flow and transport → Traded 3D geometry for a graph-based reduced system that was 500-1000 times faster in terms of computational efficiency for this specific application → Showed that the reduced-order model has great potential for providing operators with real-time decision-making information for optimizing hydrocarbon production

    • Germany
    • Automation Machinery Manufacturing
    • 700 & Above Employee
    • Graduate Intern
      • Jun 2018 - Sep 2018

      → Collaborated with a staff scientist and product manager to develop a Python code for simulating temperature evolution in additive manufacturing microslices → Collaborated with a staff scientist and product manager to develop a Python code for simulating temperature evolution in additive manufacturing microslices

Education

  • The University of Texas at Austin
    Doctor of Philosophy - PhD, Engineering Mechanics
    2012 - 2018

Community

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