Batuhan Faik Derinbay, Developer in Lausanne, Switzerland
Batuhan is available for hire
Hire Batuhan

Batuhan Faik Derinbay

Verified Expert  in Engineering

Software Engineering Developer

Location
Lausanne, Switzerland
Toptal Member Since
January 6, 2022

Batuhan is a master of computer science student at EPFL, Switzerland. He's passionate about solving real-life problems with the help of artificial intelligence. His areas of research, studies, and expertise are data science, machine learning, deep learning, computer vision, machine learning operations (MLOps), and data engineering.

Machine LearningComputer VisionPythonCC++PyTorchTensorFlowScikit-learnOpenCVPandasSQLDjangoSeleniumSoftware EngineeringArtificial Intelligence (AI)ONNX

Portfolio

Scandit
Python, C++, PyTorch, TensorFlow, Open Neural Network Exchange (ONNX)
Vestel
Python, PyTorch, Scikit-learn, Open Neural Network Exchange (ONNX), Flask
ITU Artificial Intelligence and Data Science Research Center
Python, PyTorch, TensorFlow, Keras, Pandas, Dask

Experience

Software Engineering - 3 yearsArtificial Intelligence (AI) - 3 yearsMachine Learning - 3 yearsData Science - 3 yearsDeep Learning - 3 yearsComputer Vision - 3 yearsPython - 3 yearsC++ - 3 years

Availability

Part-time

Preferred Environment

Linux, PyCharm, Slack, GitHub, Bitbucket, Jira, Confluence, Trello

The most amazing...

...product I've built is a real-time object detection and segmentation network that beat all of its commercial and non-commercial alternatives with 99% accuracy.

Work Experience

Junior Deep Learning Engineer

2020 - 2021
Scandit
  • Built a real-time universal barcode detector and decoder, using state-of-the-art shallow convolutional neural network methodologies with PyTorch and OCR for enhanced code recognition.
  • Quantized and minimized the model without accuracy sacrifices, using PyTorch and TorchScript.
  • Served the barcode detection and decoding model in real-time using FastAPI protocols with TorchServe and TensorFlow Serving.
  • Implemented HSL and RTSP protocols in C++ with PyTorch's C++ API for higher throughput capabilities.
Technologies: Python, C++, PyTorch, TensorFlow, Open Neural Network Exchange (ONNX)

Deep Learning Engineer

2020 - 2021
Vestel
  • Developed and trained a deep learning model for classifying skin images using PyTorch.
  • Deployed a segmentation model on custom depilation hardware with ONNX and served it via RESTful API.
  • Researched state-of-the-art real-time segmentation models for use in laser depilation devices.
  • Developed and trained a deep learning model for detecting hair roots on the skin using PyTorch.
  • Researched machine learning and deep learning techniques to classify images from low-level features with minimum latency.
Technologies: Python, PyTorch, Scikit-learn, Open Neural Network Exchange (ONNX), Flask

Deep Learning Researcher

2019 - 2021
ITU Artificial Intelligence and Data Science Research Center
  • Collaborated closely with several teams on industrial deep learning research projects.
  • Collected and labeled various types of tabular, image, video, and signal data from the web and designated databases.
  • Contributed to the following papers during data collection, preparation, research, and evaluation phases—"Detecting visual design principles in art and architecture through deep convolutional neural networks" and "EfficientSeg."
Technologies: Python, PyTorch, TensorFlow, Keras, Pandas, Dask

Deep Learning Engineer

2019 - 2020
Migros
  • Built an image classification model for hypermarket storehouse inventory tracking and distribution.
  • Integrated the deep learning model on internal Android and iOS applications using Java and Swift.
  • Collected many hypermarket branches' storehouse inventory data and storehouse images.
Technologies: Python, C++, Java, Swift, OpenCV, Google Cloud Platform (GCP), Android, iOS, Open Neural Network Exchange (ONNX)

Deep Learning Engineer

2018 - 2019
Bezmialem Foundation University Hospital
  • Developed software that tracks live thyroid surgeries and notifies surgeons when the parathyroid gland is present.
  • Researched and developed a state-of-the-art instance segmentation model using PyTorch and TensorFlow.
  • Built an end-to-end pipeline for real-time feedback to the surgeon using TorchServe and TensorFlow Serving.
  • Shot surgery videos for data collection, then labeled them using MATLAB video labeler.
Technologies: Python, PyTorch, MATLAB Parallel Computing Toolbox, TensorFlow

Personal Financial Assistant and Portfolio Manager

A React Native-based web and mobile app that uses several predictive machine learning models in the back end to better diversify and manage personal portfolios by individual risk assessment and personalized expenditure tracking. It is my dream to integrate artificial intelligence into asset and wealth management, and this project was one of the first steps. As the leader of a team of four, we won several fintech competitions.

Below are some news articles in Turkish that eternize our success:
• https://bit.ly/3mv7Z9x
• https://bit.ly/3Jc3z13
• https://bit.ly/3Eq37Zk

ATM Failure and Anomaly Prediction

A failure and anomaly prediction model that I built for automated teller machines (ATMs) with the data provided by the first and biggest bank in Turkey, Is Bank, during my summer internship. I also integrated a full-stack web application for better usability and user experience. Then I automated data pipelining, model updating, and deployment processes end to end. Finally, put the product in production on Is Bank's internal servers.

Languages

Python, C, C++, SQL, HTML, CSS, JavaScript, Java, Swift, GraphQL

Frameworks

Django, Selenium, Flask, Apache Spark, Scrapy, React Native

Libraries/APIs

PyTorch, TensorFlow, Scikit-learn, OpenCV, Pandas, Keras, SciPy, Natural Language Toolkit (NLTK), MLlib, Dask

Other

Machine Learning, Computer Vision, Software Engineering, Artificial Intelligence (AI), Machine Learning Operations (MLOps), Deep Learning, Scraping, MLflow, Natural Language Processing (NLP), Apache Cassandra, Sanic Web Server, Open Neural Network Exchange (ONNX), Image Processing, Sequence Models, Neural Networks, Deep Neural Networks, Chatbots, Recommendation Systems, System Design, GPT, Generative Pre-trained Transformers (GPT)

Tools

Scikit-image, Docker Swarm, Jenkins, Travis CI, PyCharm, IntelliJ IDEA, Slack, GitHub, Bitbucket, Jira, Confluence, Trello, MATLAB, MATLAB Parallel Computing Toolbox, IBM Watson, TensorBoard

Paradigms

Data Science, DevOps, Anomaly Detection

Platforms

Amazon Web Services (AWS), Google Cloud Platform (GCP), Docker, Azure, Kubernetes, Linux, Android, iOS

Storage

MongoDB, PostgreSQL

2021 - 2021

Master's Degree in Computer Science

Swiss Federal Institute of Technology Lausanne (EPFL) - Lausanne, Switzerland

2018 - 2021

Bachelor's Degree in Computer Science

Istanbul Technical University - Istanbul, Turkey

FEBRUARY 2020 - PRESENT

Applied AI Specialist

IBM

DECEMBER 2019 - PRESENT

Deep Learning Specialist

DeepLearning.AI

AUGUST 2016 - PRESENT

Machine Learning

Stanford University | via Coursera

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring