I am currently in an AI Engineer training program in Germany, working on applied machine learning and data analysis in Python. By profession I am an advertising merchant (Werbekaufmann) and a media designer for digital and print media (Mediengestalter), with 26 years of practical experience as a media designer (Adobe CC) and 30 years in sales. I work as a trainer (Ausbilder) in the Werbekaufmann profession.

Six modules of five weeks each:

1. Data Analysis with Python: Python, NumPy, pandas, Polars, Matplotlib, Seaborn, Plotly, regular expressions, web services, OOP, web scraping, time series, statistics, EDA
2. Applied Machine Learning: supervised, unsupervised and reinforcement learning, ensemble methods, decision trees, random forests, clustering, feature engineering, missing values, outlier detection, model evaluation, explainable AI
3. Deep Learning: TensorFlow, Keras, PyTorch, MLP, CNN, RNN, LSTM, GRU, autoencoders, GANs, transformers, graph neural networks, interpretability, fairness and robustness
4. NLP & Generative AI: NLTK, spaCy, TF-IDF, Word2Vec, embeddings, sentiment analysis, NER, BERT, GPT, Hugging Face, prompt engineering, LangChain, Ollama, RAG, vector databases, agentic AI, fine-tuning
5. Big Data Engineering: Apache Spark, Apache Airflow, SQL, NoSQL, ETL pipelines, data modeling and indexing
6. MLOps & Deployment: Linux, Docker, version control, MLflow, CI/CD, GitHub Actions, unit tests, FastAPI, monitoring, model deployment, portfolio

ligo-spacetime-curvature: computes the spacetime curvature of GW150914 from real LIGO strain data and shows it as a distorted grid, plus a Gravity Spy glitch analysis. Includes tests and CI.

Python, NumPy, pandas, SciPy, Matplotlib, pytest, Git, GitHub Actions

Ich absolviere derzeit eine Weiterbildung zum AI Engineer in Deutschland und arbeite mit Python an angewandtem Machine Learning und Datenanalyse. Von Beruf bin ich Werbekaufmann und Mediengestalter für Digital- und Printmedien, mit 26 Jahren Praxis als Mediengestalter (inkl. Adobe CC) und 30 Jahren im Vertrieb. Als Ausbilder bilde ich im Beruf Werbekaufmann aus.

Sechs Module à fünf Wochen:

1. Datenanalyse mit Python: Python, NumPy, pandas, Polars, Matplotlib, Seaborn, Plotly, Regular Expressions, Web Services, OOP, Web Scraping, Zeitreihen, Statistik, EDA
2. Angewandtes Maschinelles Lernen: Supervised, Unsupervised und Reinforcement Learning, Ensemble Learning, Decision Trees, Random Forests, Clustering, Feature Engineering, Missing Values, Outlier Detection, Modellbewertung, Explainable AI
3. Deep Learning: TensorFlow, Keras, PyTorch, MLP, CNN, RNN, LSTM, GRU, Autoencoder, GANs, Transformer, Graph Neural Networks, Interpretierbarkeit, Fairness und Robustheit
4. NLP & Generative AI: NLTK, spaCy, TF-IDF, Word2Vec, Embeddings, Sentiment Analysis, NER, BERT, GPT, Hugging Face, Prompt Engineering, LangChain, Ollama, RAG, Vektordatenbanken, Agentic AI, Fine-Tuning
5. Big Data Engineering: Apache Spark, Apache Airflow, SQL, NoSQL, ETL/Pipelines, Data Modeling und Indexing
6. MLOps & Deployment: Linux, Docker, Version Control, MLflow, CI/CD, GitHub Actions, Unit Tests, FastAPI, Monitoring, Model Deployment, Portfolio

ligo-spacetime-curvature: berechnet die Raumzeit-Krümmung von GW150914 aus echten LIGO-Strain-Daten und stellt sie als verzerrtes Gitter dar, dazu eine Glitch-Analyse mit Gravity Spy. Mit Tests und CI.

Python, NumPy, pandas, SciPy, Matplotlib, pytest, Git, GitHub Actions