"A rising star who has spent the last several years building advanced ML, computer vision, and 3D capabilities. Sudarsna also brings strong front-end experience and a track record of diving deep into complex technical challenges" - Todd Hagen, Principal Engineer

  • Education

  • MS, Computer Science

    University of Minnesota

    BS, Information Technology

    Northeastern University CPS

    Graduated cum laude

  • Work Experience

  • Masters Research Assistant

    Interactive Visualization Lab

    Research assistant on motion tracking projects using Touchdesigner, depth estimation, and computer vision.

    Interested and working in the intersectionality of dance and computer science.

  • Software Engineer

    Target (2022 - Present)

    Developed a React.js endpoint to deliver AI powered content moderation to internal users at Target.

    Created a codebase to calibrate 9 cameras using OpenCV and knowledge of camera calibration mathematics. Used these calibrations to successfully triangulate, and 3D model people walking in space from the 9 camera angles.

    Restructured machine learning outfit recommendation code to modularize score generation and random forest processes

    Consolidated room-planner and furniture workflows to improve code maintainability and wrote a pyspark function to filter out 3D assets

    Conducted a literature review and comparative analysis of best methods for 3D mesh metric quality analysis and 3D mesh patching.

    Wrote a camera detection handler C++ script for a store robotic camera

    Worked on server migration tasks and re-architecture for a machine learning outfit completion and recommendation application

    Trained NeRF algorithm with custom 2D dataset and COLMAP image processing on Kubernetes clusters to create 3D assets

    Participated in Target’s internal hackathon and their internal research and idea development program

    Adapted code for machine learning smoothing algorithms and developed new code to transform and rotate meshes back to their original center

    Developed scripts to make mesh objects manifold and to upsample meshes to a certain level of resolution

    Architected ROS2 solution to read price tags from store shelves using YOLOv8 object detection and EasyOCR character recognition

  • Transportation Pathways Intern

    Federal Highway Administration (2020-2021)

    Created workflows for financial data using tools like SAP Business Objects, PowerBI, SharePoint, Outlook

    Participated in formal review of business and technical requirements and application design, including Section 508 reviews for accessibility and NEPA/EIS reviews

    Assisted with developing solutions pertaining to community transportation equity

    Developed SQL scripts to query financial databases and presented historical data trends to the team

  • Web Developer

    Giardina and French (2021)

    Demonstrated knowledge of quality assurance testing methodology relative to Agile methodology

    Worked directly with non-technical users for design, testing and workflow