01.04.2026 aktualisiert
verifiziert
Premiumkunde
100 % verfügbarMachine Learning Engineer | AI Engineer | Data Scientist | Statistician | Team-Lead | Python
Eichenau, Deutschland
Deutschland +2
M.Sc. Statistikinfo: Deutschland, Österreich, Schweiz
Über mich
Als erfahrener ML-Engineer bringe ich Ihre ML-Projekte End-to-End in Produktion. Ich entwickle effiziente, skalierbare KI-Lösungen, treibe Innovationen voran und setze Technologie gezielt ein, um messbaren Geschäftswert zu schaffen.
Skills
Künstliche IntelligenzComputer VisionBash ShellKreative ProblemlösungPythonMachine LearningAzure Machine LearningData SciencePyTorchLarge Language ModelsPrompt EngineeringDeep LearningFastAPIPySparkScikit-learn
Driven by Business - Powered by Tech!
Business-Ziele und Ideen stehen an erster Stelle – Technologie macht sie greifbar und umsetzbar.
Als erfahrener ML-Engineer bringe ich ihre ML Projekte End-to-End in Produktion.
Angefangen bei der Datenanbindung, der effektiven Transformation mit Python und Pyspark, Modelloptimierung mit Scikit-Learn, PyTorch und Pyro, dem Deployment via IaC, Docker und Kubernetes, und dem Monitoring der Ergebnisse via MLOps-Tools wie z.B. Databricks-MLFlow.
Darüber hinaus konnte ich bereits kundengetriebene Erfahrung in der Leitung von Data-Science-Teams und dem Wissensaufbau zum Thema ML und AI sammeln.
Experten Skills (8+ Jahre Berufserfahrung):
- Data Analysis, Machine Learning mit Python (8 + Jahre)
- Deep Learning, Computer Vision mit Python (8 + Jahre)
- Prompt Engineering, RAG-Implementation and Agentic AI mit Python (4 + Jahre)
- Agiles Software Development (8+ Jahre)
- ML-Ops und Data-Pipelines in Cloudumgebungen(6 + Jahre)
- Management von Data-Science-Teams (4+ Jahre)
- Knowledge Sharing (Coding Workshops, Courses) (7 + Jahre)
Unterstützende Skills:
- Unit- und E2E-testing (6 + Jahre)
- Docker und Kubernetes (6 + Jahre)
Sprachen
DeutschMutterspracheEnglischverhandlungssicher
Projekthistorie
This initiative represents a cutting-edge data science project aimed at crafting a scalable, robust, and highly accurate demand estimation model for better resource and delivery planning. Harnessing the full potential of Azure Cloud's advanced capabilities, the project employs state-of-the-art machine learning algorithms, Sentiment Analysis via LLMs, Model-Explanations via Dash and GenAI-Agents, large-scale data processing pipelines and cloud-based infrastructure to forecast demand across all business units. By optimizing production and logistics, this model drives efficiency and elevate decision-making.
- Creation of an end-to-end workflow tailored to accurately predict demand patterns and behaviors for all business units
- Two-Pronged Approach for Intermittent Demand Forecasting
▪ Binary Prediction: Identifying the occurrence of the next demand instance.
▪ Regression: Estimating the magnitude of the demand.
- Top-Down Forecast pronged together with Bottom-Up-Forecast to increase performance
- Implementation ensemble techniques.
- Use of Azure Synapse for streamlined scalability.
- Leveraging Large Language Models (LLMs) and Azure OpenAI to implement sentiment analysis for customer reports, enhancing qualitative insights.
- Implementation of a GenAI-Agent-System to use data and model logs to explain model outcome
- Deployment of the GenAI-Agent-System as plotly-Dash-App via Docker and Kubernetes
- Deployment of MLFlow-Tracking-Server via Docker and Kubernetes
This initiative represents a state-of-the-art data science project focused on building a scalable, probabilistic, and highly reliable cashflow forecasting system to empower strategic financial planning. Leveraging the advanced capabilities of Azure Cloud, the project integrates Bayesian modeling with Pyro, cloud-native data pipelines, and automated MLOps workflows to generate accurate, uncertainty-aware cashflow projections across all business domains. By combining probabilistic programming, large-scale data processing, and end-to-end Azure infrastructure, the solution delivers transparent, quantifiable forecasts that enable finance teams to anticipate liquidity needs and optimize capital allocation.
Scope of Work:
- Creation of an end-to-end workflow tailored to accurately predict cashflow patterns via Pipelines and APIs with FastAPI
- Two-Pronged Approach for Receivables and Payables with Pyro
- Cashflow-Prediction as combination of Two-Pronged Approach for Receivables and Payables with Pyro
- Implementation ensemble techniques to bolster prediction precision.
- Evaluate Feature Importance with Shapley Values
- Explain Feature Importance with an Agentic AI-Explainer
- Use of Azure Synapse for streamlined scalability and accessibility.
- Deployment of MLFlow-Tracking-Server via Docker and Kubernetes
This initiative represents a cutting-edge data science project aimed at crafting a scalable, robust, and highly accurate demand estimation model for better resource and delivery planning. Harnessing the full potential of Azure Cloud's advanced capabilities, the project employs state-of-the-art machine learning algorithms, Sentiment Analysis via LLMs, Model-Explanations via Dash and GenAI-Agents, large-scale data processing pipelines and cloud-based infrastructure to forecast demand across all business units. By optimizing production and logistics, this model drives efficiency and elevate decision-making.
- Creation of an end-to-end workflow tailored to accurately predict demand patterns and behaviors for all business units
- Two-Pronged Approach for Intermittent Demand Forecasting
▪ Binary Prediction: Identifying the occurrence of the next demand instance.
▪ Regression: Estimating the magnitude of the demand.
- Top-Down Forecast pronged together with Bottom-Up-Forecast to increase performance
- Implementation ensemble techniques.
- Use of Azure Synapse for streamlined scalability.
- Leveraging Large Language Models (LLMs) and Azure OpenAI to implement sentiment analysis for customer reports, enhancing qualitative insights.
- Implementation of a GenAI-Agent-System to use data and model logs to explain model outcome
- Deployment of the GenAI-Agent-System as plotly-Dash-App via Docker and Kubernetes
- Deployment of MLFlow-Tracking-Server via Docker and Kubernetes
Portfolio
MLOps Retraining
Automatic Retraining Workflow done with Azure
Boardgame Copilot
Android GenAI App that uses boardgame-rules via langchain and openai to help understand the rules and ask questions.
mlflow overview
Overview of training runs tracked in mlflow
training plots
training plots to evaluate model results in azure
azure-ml overview
Overview of training runs in azure-ml
exali Berufshaftpflicht-Siegel
Das original exali Berufshaftpflicht-Siegel bestätigt dem Auftraggeber, dass die betreffende Person oder Firma eine aktuell gültige branchenspezifische Berufs- bzw. Betriebshaftpflichtversicherung abgeschlossen hat.
Versichert bis: 01.01.2027