Hi, I'm Nik

I enjoy connecting product thinking with data and coding — this portfolio is where I share some of the projects I’ve built along the way.

SENTIMENT ANALYSIS, PUBLICATION

Understanding Player Feedback on Game Updates

I conducted a sentiment analysis of League of Legends patch-note discussions on Reddit, analyzing 14,500+ comments with the Reddit API and Google Cloud NLP. The project highlights how update-labeled threads enable patch-specific insights, revealing consistently negative player reactions and pinpointing balance concerns and the controversial “Vanguard” anti-cheat system. This work forms the basis of an academic publication, addressing a gap in game analytics research.

• View on Kaggle
• Publication Q4/2025

PYTHON, API

Automating Therapist Search from Germany’s 116117 Service

I developed a Streamlit app in Python that connects to Germany’s 116117 healthcare service, automating therapist searches by zip code and filters. The app extracts and structures the results into a standardized dataset, applies preprocessing for clean exports, and generates an automated call schedule. By replacing manual copy-paste with an API-driven workflow, it demonstrates how data engineering can simplify real-world healthcare access.Update Q3/2025: The official developers of the 116117 app have confirmed that they will integrate the app’s core functionality into Germany's medical appointment service.

• View on Github
• View Webapp

DATA SCIENCE, MACHINE LEARNING

Predictive maintenance of machines with ML

I created a predictive model to identify machine failures, achieving 88% accuracy in predicting machine failures to minimize downtime and costs. Implemented using machine learning techniques with a focus on high recall to capture potential failures effectively. Coded in Python.

• View on Kaggle
• View on GitHub

PYTHON, API

Automated Job Data Fetcher

I created a script that fetches new job postings from the Bundesagentur für Arbeit website containing the keyword 'data' and exports all relevant job details, such as the company and job description. The script runs daily through a GitHub Actions workflow to automate the data collection process. Based on this data, extensive analysis can be conducted.

• View on GitHub

DATA ENGINEERING, BACKEND

Real-time data backend using Docker, Kafka & MongoDB

I developed a simplified data back-end of a football player tracking solution, enabling analyses such as formation changes over time. Technical implementation using Docker for containerization of microservices, Kafka for data-streaming and MongoDB for storage.

• View on GitHub

BACHELOR THESIS

Data-driven Sales with Consumer Goods Registration Data

For my bachelor thesis at Bosch Power Tools, I developed a concept of a data-driven sales tool that was later adopted by the company. This solution offers the sales team a unified view of customer purchases, with insights across companies and product lines.

• On request

© Niklas Böckmann. All right reserved.