Summary
Overview
Work History
Education
Skills
Nationality
Accomplishments
Languages
Affiliations
Websites
Timeline
Generic
Souvik Tewary

Souvik Tewary

Potsdam

Summary

Seasoned Machine Learning Engineer with 3+ years of experience from industry and institute. Highly accomplished in digital signal processing (audio processing, feature extraction) from my last industrial experience. Demonstrated planning, development and optimization in multiple projects including publication in the field bio-telemetry system, image sagmentation software (MATLAB). Highly motivated team player and organizational goal oriented.

Overview

6
6
years of professional experience

Work History

Fulltime Job

Diametos GmbH
Potsdam
01.2022 - 07.2023
  • Development, cleaning and pre-processing of RAW audio data
  • Optimization of pre-existing snoring-detection algorithm
  • research, test and implementation of novel signal processing and feature extraction algorithm
  • Technical documentation according to the regulatory standards for medical device
  • statistical model (ML approach) for detecting different levels of snoring type based on the acquired features. (SVM; AdaBoost, XG Boost)
  • Planning and conception to project coordination. Docker hub, JIra. Following agile methodology for the whole project.

Master Thesis

Friedrich-Alexander-Universität (FAU)
Erlangen
02.2020 - 08.2020
  • Automatic Assessment of Parkinson's Disease Using Different Speech Dimensions and AdaBoost Classifier
  • Using Supervised Machine Learning.
  • Input Parameter: Voice data from 100 subjects (University of Columbia)
  • Preprocessing the audio data for feature extraction using Disvoice, praat.
  • Building statistical model (AdaBoost) model from scratch
  • Libraries: Scikit-Learn
  • Prediction Analysis by applying 50 new test subjects to the trained AdaBoost model
  • Final test accuracy: 80%

Teaching Assistant

Pattern Recognition Subject
10.2019 - 03.2020
  • Assisting students with laboratory experiments in python
  • Experiments include Learning of the basic algorithm for classification and regressor problems, such as Bayesian, Gaussian Mixture Model Expectation Maximization, SVM
  • Checking and grading the students based on their python submission after the exercise

Research Internship

Adidas AG & M.A.D LAB (FAU)
Erlangen
10.2017 - 04.2018
  • Smart and Wearable shoe for runners using energy harvesters
  • Developing a shoe that can count steps by using energy harvesters as an energy source
  • Energy-harvester input: flexi solar panel, piezoelectric sensor and electromagnetic induction were used to generate energy
  • Using AutoCAD to design the electromagnetic induction
  • microcontroller with bluetooth 4.0 for detecting and sending step-count to the frontend ("smart-shoe" app).

Research assistant

Department of Computer Science (FAU)
Erlangen
02.2017 - 08.2017
  • Recognition and classification of digital histological images
  • Separation of the chorioallantoic membrane CAM of the chick embryo from the Tumor
  • Segmenting fibroblasts that have migrated from the CAM into the tumor.

Education

M.Sc in Medizintechnik - Medical Imaging and Data Processing

Universität Erlangen-Nürnberg (FAU)
04.2021

B.Tech in Biomedical Engineering -

West Bengal University of Technology (WBUT)
06.2014

Skills

  • Machine Learning, Statistics, Deep Learning (tensorflow, keras), AI audio analysis, data visualization and presentation, predictive analysis, Agile methodology (SCRUM), DevOps (Git, Docker), OpenCV, Pytorch, numpy, pandas, scikit-learn, LabView, MATLAB

Nationality

Indian

Accomplishments

  • A Novel Approach towards designing a Wearable Smart Health Monitoring System measuring the vital parameters in emergency situations in Real-Time and providing the necessary Medical Care through Telemedicine. https://ieeexplore.ieee.org/document/7509332
  • An Automated Graphical User Interface based System for the Extraction of Retinal Blood Vessels using Kirsch‘s Template. https://thesai.org/Publications/ViewPaper?Volume=6&Issue=6&Code=IJACSA&SerialNo=13
  • several national Awards in robotics building and task forming events.

Languages

Bengali
First Language
English
Proficient (C2)
C2
German
Intermediate (B1)
B1

Affiliations

  • music enthusiastic, cricket, pub crawler.

Timeline

Fulltime Job

Diametos GmbH
01.2022 - 07.2023

Master Thesis

Friedrich-Alexander-Universität (FAU)
02.2020 - 08.2020

Teaching Assistant

Pattern Recognition Subject
10.2019 - 03.2020

Research Internship

Adidas AG & M.A.D LAB (FAU)
10.2017 - 04.2018

Research assistant

Department of Computer Science (FAU)
02.2017 - 08.2017

M.Sc in Medizintechnik - Medical Imaging and Data Processing

Universität Erlangen-Nürnberg (FAU)

B.Tech in Biomedical Engineering -

West Bengal University of Technology (WBUT)
Souvik Tewary