About Me

I'm a generative AI architect from Germany. I'm mostly interested in creating tangible value for people and businesses through data and AI using the most effective and not the fanciest technology. As you can see below I like working with data and tech from various fields and apply them in different ML, AI, and data analytics projects.
Most of my projects are with Python and associated libraries while I'm deepening my skills at SQL, Tableau.

I've studied computer science and information systems with focus on data science in my bachelor and master. I have been working as a Business and Solution Architect at on several enterprise project for more than 6 years. My major topics have been data platforms, Generative AI and Trustworthy AI.

Portfolio

Foto von Growtika auf Unsplash

Personal Household Agent

A personal assistant via Telegram to manage telegram and home assistant.

Langchain, MCP, FastAPI, Prompt Engineering

Source Code
Foto von Fionn Große auf Unsplash

German Election Programs Analysis

Analysis of election programs with Natural Language Processing and Development of local LLM-based Chatbot.

NLP, RAG, Spacy, Langchain, Vectorstore

Reports Github repo
Dashboard, picture by Philip Singer

Dashboard of Sustainable Car Use

Data visualization of official car license data in Germany with focus on alternative drives.

Data visualization, Tableau, Open Data, Data Storytelling

Tableau Dashboard
By Polina Tankilevitch

Clustering chocolate -
failed project

Gathering and clustering open U.S. chocolate data, because customer segementation is boring...

KMeans, DBSCAN, unsupervised, REST API, PostgreSQL

Source Code here
Forecast, picture by Philip Singer

Cyclists Forecasting

Data augmentation with Python and forecasting using Prophet algorithm. Explain the model with Shapley values.

Time-series forecast, Prophet, Streamlit, SHAP

Source Code here Streamlit Web App
Foto von Cristiana Raluca von Pexels

Berlin bike use analysis

Data exploration of bike count stations in Berlin. Prepring data and views with SQL and visualization with Tableau.

Data Exploration, SQL, Tableau

Source Code here Tableau Dashboard
Image by author

Architecture for real-time Data Preparation

Combine and Preprocess Your Heterogeneous Data for Analytics with Apache Flink

Data Pipelining, Apache Flink, real-time

Read Medium article

Telco customer churn Prediction

Predict whether telco customers will leave the company: Exploratory Data Analysis, Machine Learning and Deep Learning

Classification, Scikit-Learn, Keras, Pandas

See source code Learn more soon

House pricing prediction

CRISP-DM Data Science Project in Python: Exploratory Data Analysis and Machine Learning

Regression, Scikit-Learn, Pandas

source code here