CONTACT

goksinan@gmail.com

github-goksinan

linkedin-sinangok

SKILLS

TOOLS

Python, C++, MATLAB, AWS, Git, GitHub, CMake, PyCharm, VS Code, OpenCV, Dlib, LLMs.

ML

PyTorch, Scikit-learn, Numpy, Scipy, Pandas, Timm, Albumentations, PyTorch Metric Learning, Weights&Biases.

ML DEPLOYMENT

Onnxruntime, Libtorch, OpenVINO, MNN, Apache TVM, ML Flow, Sagemaker, Kubernetes.

APPLIED ML

Computer Vision, Deep Learning, Classification, Linear Regression, Object Detection, Landmark Detection, Anomaly Detection, Similarity Learning, Contrastive Learning.

EDUCATION

PhD in BIOMEDICAL ENGINEERING

Rutgers University & New Jersey Institute of Technology

2012-2018

MS in ELECTRICAL ENGINEERING

Lehigh University

2009-2011

BS in ELECTRICAL AND COMMUNICATIONS ENGINEERING

Kocaeli University, Turkey

2002-2006

ABOUT

ML Engineer with over 5 years of experience in data collection, analysis, and visualization. In-depth knowledge of digital signal processing, image processing, machine learning and computer vision. Expertise in training and deploying deep learning models for production.

WORK EXPERIENCE

MACHINE LEARNING ENGINEER

Iris ID Systems, Inc. | 2020 - current

Developed and deployed machine learning models to be used in various stages of the biometric identification.

  • Significantly improved the identification accuracy (x2) by developing a model from scratch.
  • Developed a new countermeasure algorithm and reduced the false positives by ~90%
  • Created novel face anti-spoofing algorithms to detect 2D and 3D presentation attacks.
  • Enhanced face detection accuracy, including for faces with masks and PPE, by developing an advanced object detection model.
  • Helped carrying the company to a prestigious 3rd position in National Institute of Standards and Technology’s (NIST) Irex10 competition.

ADJUNCT FACULTY

The College of New Jersey | Spring 2015, Fall 2019

  • Instructed senior-level course on Bioinstrumentation, focusing on biomedical signal processing techniques for EMG, EEG, and ECG data.
  • Taught the Physiological Systems-I course, integrating data analysis and statistical methods to interpret physiological signals.
  • Utilized statistical software and programming languages (e.g., Python, MATLAB) to teach students data analysis and visualization techniques.

New Jersey Institute of Technology | Fall 2019

  • Instructed the Fundamentals of Engineering Design course, teaching students the basics of programming, design, and engineering.

DATA SCIENTIST

New Jersey Institute of Technology | 2018 - 2019

  • Led multiple research projects, overseeing data collection, analysis, and interpretation to drive insights and inform decision-making.
  • Utilized machine learning techniques to analyze complex biomedical data, identifying patterns and predicting outcomes.
  • Assisted in receiving a research grant ($300K) funded by the National Institute of Health (NIH).

RESEARCH ENGINEER

New Jersey Institute of Technology | 2012 - 2018

  • Designed and built a data acquisition system that fully automated the data collection process.
  • Utilized ML techniques (classification and regression) to investigate the neural control of movement.
  • Published research papers and attended academic conferences.

SELECTED PUBLICATIONS

  • “Prediction of forelimb EMGs and movement phases from corticospinal signals in the rat during the reach-to-pull task,” International Journal of Neural Systems, 2019. Link

  • “Convolutional networks outperform linear decoders in predicting EMG from spinal cord signals,” Frontiers in Neuroscience, 2018. Link

  • Google Scholar

    ResearchGate

    HONORS AND AWARDS

    RESEARCH TRAINING GRANT | 2019

    Received a training grant from New Jersey Commission on Spinal Cord Research regarding spinal cord injury research.

    PROGRAM SELECTEE | 2017

    Selected to participate in the second NIH-sponsored Summer School in Adaptive Neurotechnologies.

    GRADUATE SCHOLARSHIP | 2008

    Full scholarship awarded by Turkish Ministry of National Education for ESL and graduate education in the US.