Machine Learning Engineer
A Data Scientist and software engineer with 4+ years of industry experience in software development and data analytics. I have developed extensive knowledge and expertise in the data science domain, automation, and analytics by working at Samsung, Fraunhofer Fokus, and Dreamlines GmbH. During the course of my work, I get my hands dirty on collecting and examining large datasets. I developed a product recommendation tool using variational autoencoder in real-time. This helps us in generating 1.5% additional bookings. Furthermore, I worked on a newsletter tool that uses the XGBoost framework which turned out to be a handy tool to let the marketing team know, the next likely booking dates for users. I made sure that both these models should work in production smoothly by creating rest end-points, Docker containers and regular build using Jenkins. In addition to that, I am competent in creating and implementing forecasting models, producing detail reports and communicating the results across the team. My previous roles have helped me to develop and significantly improve my attention to detail and accuracy and gave me a chance to know-how what it takes to work in a deadline-driven business environment.
In addition to my experience, I can offer the organization excellent analytical skills, important multitasking, and time management abilities. This is also backed up with my through educational and research background, with a bachelor in Computer Science followed by the master(M.Sc) in Computer Science from the Technische Universität Berlin with a major in Data & Software Engineering and a GPA of 1.6. During my master thesis with the title "Spatial & Temporal feature extraction for gaming Video Quality Assessment(VQA)", I implemented a system that does the feature extraction of a given video for the visualization of video data. It also involved the modeling of the video data and the predictive analysis using state-of-art Machine Learning algorithms and popular libraries & tools such as Scikit-learn, Scikit-video, PySpark, Spark ML and Pandas. In the end, I built a lightweight No-Reference(NR) metric, named BEGVQ for VQA of gaming videos, which outperforms the state-of-the-art NR metrics.
Skilled in Python, Pandas, Scikit-Learn, PySpark, SQL, Machine Learning, Deep Learning(Autoencoders, CNN), and Data Analysis, Git, Jenkins, Docker, AWS(Basic), NoSql(Redis), Seaborn, Matplotlib, Flask.
Work in Germany
Expected Base Salary
Expected Hourly Rate
Total Professional Experience
Send a connection request to the candidate to get their contact details.Contact Candidate