Summary
- I am a Senior Software and ML Engineer at Aiva Technologies, working on algorithmic music generation with generative AI for x00k users, including clients like Nvidia, Steve Woz, and Vodafone.
- In my spare time, I develop and refine algorithmic models for predictive analytics and trading in quantitative finance and algorithmic sports betting. This has given me valuable experience in analyzing large datasets, implementing statistical models, and developing software for automated trading systems.
Skills
- Programming Languages: Python, Rust, C++
- Deep Learning Frameworks: Keras, Pytorch
- Databases: MongoDB, PostgreSQL, DuckDB
- ML Libraries: scikit-learn, Numpy, Prophet, OpenCV, GluonTS, Neural Forecast, statsmodels
- Programming Languages: Python, Rust, C++
- Data Related: Spark, Arrow, Pydantic, peewee, Airflow, Pandas/Polars, Luigi, SQLModel
Work Experiences
Senior Machine Learning and Software Engineer at AIVA Technologies
From January 2019
- Joined as an intern and rapidly advanced to a leadership position within the backend development team. Played a pivotal role in taking the platform from its pre-alpha stage to a profitable state, leveraging a combination of technical expertise, project management skills, and strategic vision.
- Assumed responsibility for the Python backend development, overseeing the design and implementation of a sophisticated system that utilizes a combination of machine learning and deep learning-based algorithms to generate musical compositions. The system has proven to be highly effective, serving more than 1 million AI-generated songs per month.
- Developed numerous Extract, Transform, and Load (ETL) pipelines with the purpose of extracting, organizing, and cleansing data to facilitate the building of effective machine learning models.
- Developed and deployed internal tools for team members to automate data quality control and validation, leveraging libraries such as pydantic and Jupyter notebooks.
Research Assistant at AudioLabs
From January 2020 to September 2021
- Designed and developed machine learning-based algorithms for singing voice detection, which are widely used as a pre-processing tool for many internal research projects. These algorithms were implemented using the Keras and Scikit-Learn libraries.
Summer Research Intern at Fraunhofer IAIS
From June 2018 to September 2018
- Evaluated and tested several cutting-edge algorithms for speech segmentation, ultimately proposing a novel method for a government-funded project. This new method was implemented and has been adopted by at least 10 internal projects as a result of its demonstrated effectiveness.
Summer Research Intern at Telecom ParisTech
From June 2017 to September 2017
- Designed and developed a novel probabilistic model based on α-Stable distributions, along with an inference algorithm for parameter estimation. The algorithm and the overall approach were implemented and tested on a real speech enhancement problem, which demonstrated superior performance compared to several state-of-the-art models across various tasks.
Teaching Assistant at Sabanci University
From September 2017 to January 2019
- Held weekly recitation hours and office hours for Introduction to Probability Course.
Education
MSc. in Communication and Multimedia Engineering from Friedrich-Alexander-Universitat Erlangen-Nurnberg
From September 2019 to January 2022
- Thesis Title: Data-Driven Singing Activity Detection for Music Recordings
- GPA: 1.9/1.0
BSc. in Electronics Engineering from Sabanci University
From September 2014 to January 2019
- GPA: 3.78/4.0
- High Honour for Every Term