Konstantine Tsafatinos

Full Stack Developer at Neuromatch
  • Claim this Profile
Contact Information
Location
Canada, CA

Topline Score

Bio

Generated by
Topline AI

0

/5.0
/ Based on 0 ratings
  • (0)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

No reviews to display There are currently no reviews available.

0

/5.0
/ Based on 0 ratings
  • (0)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

No reviews to display There are currently no reviews available.
You need to have a working account to view this content. Click here to join now

Experience

    • United States
    • Non-profit Organizations
    • 1 - 100 Employee
    • Full Stack Developer
      • Feb 2023 - Present

      Toronto, Ontario, Canada Acting as the go-to tech guy for everything from systems admin, to data management, to frontend/backend development, to domain/network management and product development. Leading a team of volunteers.

    • Real Estate
    • 1 - 100 Employee
    • Digital Wizard
      • Aug 2021 - Present

      Toronto, Ontario, Canada

    • Canada
    • Events Services
    • 1 - 100 Employee
    • Digital Consultant
      • May 2022 - Dec 2022

      Toronto, Ontario, Canada

    • Canada
    • Staffing and Recruiting
    • 1 - 100 Employee
    • Contract Data Scientist
      • Aug 2021 - Jan 2022

      Toronto, Ontario, Canada Conducted a review of their personality profiling product; visualized distributions of product usage data using normalized histograms, performed Principle Component Analysis and k-means clustering (The Elbow Method, Silhouette Analysis and Gap Statistics) to assess and visualize profile types in reduced dimensional space, provided insights based on observed patterns and suggested improvements to their product.

    • United States
    • Civic and Social Organizations
    • Student Researcher
      • Feb 2020 - Feb 2021

      Germany Performed analysis of psychiatric data using Random Forests and Feature Importance (Gini importance and SHAP values); enabling health care professionals to predict the efficacy of deep brain stimulation treatment, used to treat patients with severe Parkinson's disease, and providing them with metrics to consider the contribution of input variables to each prediction, improving model interpretability. Taught an intro to Machine Learning course offered to Charité staff as well as conducted… Show more Performed analysis of psychiatric data using Random Forests and Feature Importance (Gini importance and SHAP values); enabling health care professionals to predict the efficacy of deep brain stimulation treatment, used to treat patients with severe Parkinson's disease, and providing them with metrics to consider the contribution of input variables to each prediction, improving model interpretability. Taught an intro to Machine Learning course offered to Charité staff as well as conducted research into the effects of noise on classification of MRIs. Show less

    • Canada
    • Hospitals and Health Care
    • 1 - 100 Employee
    • Associate Tech Analyst
      • May 2018 - May 2019

      Toronto, Canada Area Created gesture recognition software in Python with the aim of developing a non-contact human computer interface, saving surgeons time by allowing them to review scans without having to unscrub. Responsible for the design, development and application of a pipeline that incorporated machine learning models for collected training, validation, and test sets for different hand gestures. Random Forests, Multi-layer Perceptrons, Support Vector Machines and other algorithms were implemented in an… Show more Created gesture recognition software in Python with the aim of developing a non-contact human computer interface, saving surgeons time by allowing them to review scans without having to unscrub. Responsible for the design, development and application of a pipeline that incorporated machine learning models for collected training, validation, and test sets for different hand gestures. Random Forests, Multi-layer Perceptrons, Support Vector Machines and other algorithms were implemented in an ensemble model to improve the accuracy (to 99.5%) and generalization of the previously existing hard coded gesture recognition software. Show less

    • Canada
    • Software Development
    • 1 - 100 Employee
    • Javascript Developer
      • Apr 2018 - May 2018

      Toronto, Canada Area Fixed application issues flagged by users and developers; fixed a soft delete bug, removed the ability to re-open past appointments, added color coded buttons improving the user interface, etc.

Education

  • Technische Universität Berlin
    Master's degree, Computational Neuroscience
    2019 - 2024
  • University of Waterloo
    Bachelor of Applied Science (B.A.Sc.), Civil Engineering
    2006 - 2011

Community

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