07.10.2026 aktualisiert
100 % verfügbar
AI Engineer | Machine Learning (ML) & Agentic AI | GenAI, Databricks, Mosaic AI, GCP, LangChain, AI
Berlin, Deutschland
Weltweit
(M.Sc.) Master of Science in Data Science Über mich
ML & AI Engineer with 6 years of Python/ML experience. Specialized in Agentic AI, GenAI, Deep Learning (DL), Machine Learning (ML), Databricks & LangChain. MSc Data Science. Former EY consultant with experience leading AI & ML projects. Additional expertise in Football & Sports AI and Analytics.
Skills
AI AgentsRAGMLflowLLMMLOpsGCPVertex AILangChainKünstliche IntelligenzKünstliche Neurale NetzwerkeAtlassian JiraGitHubPythonMachine LearningNumPy
AI Engineer, Machine Learning Engineer, ML Engineer, Generative AI Engineer, GenAI Engineer, Agentic AI Engineer, Data Scientist, Reinforment Learning Engineer, RL Engineer, Deep Learning Engineer, DL Engineer, AI Consultant, Agentic AI Consultant, GenAI Consultant, RL Consultant, DL Consultant, ML Consultant, Machine Learning, Deep Learning, Reinforcement Learning, Generative AI, GenAI, Agentic AI, Artificial Intelligence, AgentOps, Agent Operations, LLMOps, LLM Operations, GenAIOps, GenAI Operations, RLOps, Reinforcement Learning Operations, DLOps, Deep Leardning Operations, MLOps, Machine Learning Operations, Large Language Models, LLM, LLM Agents, AI Agents, Agentic Systems, Multi-Agentic Systems, RAG, Retrieval Augmented Generation, AI Context Engineering, Context Engineering, Prompt Engineering, Tool Use, Agent Evaluation, Model Evaluation, Model Calibration, Feature Engineering, Spark MLlib, AutoML, Artificial Neural Networks, Neural Networks, NNs, Convolutional Neural Networks, CNN, Recurrent Neural Networks, RNN, Transformers, Attention Mechanism, Graph Neural Networks, GNN, Heterogeneous Graph Transformers, HGT, Python, PySpark, SQL, PyTorch, PyTorch Geometric, PyG, scikit learn, scikit-learn, Keras, NumPy, pandas, Matplotlib, Seaborn, MLflow, LangChain, LangGraph, LangSmith, Databricks, Mosaic AI, Databricks Agent Bricks, Databricks Genie Agents, Databricks Apps, Google Cloud Platform, GCP, Vertex AI, Gemini Enterprise Agent Platform, Google Colaboratory, Google Colab, BigQuery, Dataform, Azure ML, Azure ML Studio, Jupyer Lab, Jupyter Notebook, Visual Studio Code, VS Code, Power BI, Tableau, Streamlit, FastAPI, Atlassian Jira, Jira, Git, GitHub, Data Analytics, AI Analytics, Sports Analytics, Sports AI, Football Analytics, Football AI, Sports Data Science, Football Data Science, Event Data, Tracking Data, Expected Goals, xG, Expected Possession Value, EPV, Expected Possession Value-Added, EPV-Added, Spearman Pitch Control, Pitch Control, Player Scouting, Performance Analytics, Tactical Data Analytics, Football Modelling, Football Modeling, Sports Modelling, Sports Modeling, Physics-Based Modelling, Physics-Based Modeling.
Erfahrung
Berufserfahrung0 - 5 Jahre
Freelancer seit2026
Sprachen
DeutschGrundkenntnisseEnglischverhandlungssicherGriechischgutItalienischGrundkenntnisseSpanischMuttersprache
Projekthistorie
- Selected onto Twelve Football's 1st edition of their “AI Context Engineering for Football” course -a very limited cohort of 45 participants worldwide, hand-picked by Founder David Sumpter himself plus his committee- and delivered the accompanying Agentic AI Engineering project as sole developer.
- Designed, developed, and delivered a conversational AI Analyst agent for Centre-Backs (CBs) & CB-pairings evaluation & scouting, grounding an LLM in structured football data through AI context engineering and tool-use techniques.
- Released as a fully equipped & governed Streamlit front-end chat application, answering natural-language scouting questions with analyst-grade player analyses, comparisons, pairing recommendations, and more...
TECH STACK: Python · LLM tool use & context engineering · RAG · LangChain / LangGraph / LangSmith · scikit-learn · NumPy · pandas · Matplotlib · Seaborn · football event data · Streamlit · UI/UX · GitHub · Git · VS Code
- Designed and built a new Sports AI & Analytics business line from scratch – single-handedly and as a parallel initiative alongside daily delivery – after proactively pitching the idea & business opportunity to the entire department, securing sponsorship and direct supervision from a department-linked share-holding Partner.
- Delivered end-to-end Machine Learning (ML) and Deep Learning (DL) solutions for clients: data preparation and feature engineering, model training and evaluation, experiment tracking, MLOps, deployment and monitoring.
- Developed Generative (GenAI) & Agentic AI solutions for clients: LLM-based knowledge assistants, retrieval-augmented (RAGs), and tool-using agents.
- Led a small team of 3 technical people: task planning, technical guidance, code review, and delivery quality.
- Built decision-ready dashboards and visual analytics: translating model output into recommendations for business stakeholders.
- Ran cross-functional collaboration with technical and non-technical client teams, from requirements workshops through to hand-over.
TECH STACK: Python · PySpark · SQL · Databricks (Mosaic AI, Spark MLlib, MLflow, AutoML, Agent Bricks, Genie Agents, Databricks Apps) · GCP (Gemini Enterprise Agent Platform/Vertex AI, BigQuery, Dataform) · Azure ML Studio · LangChain / LangGraph / LangSmith · scikit-learn · PyTorch · Keras · NumPy · pandas · Matplotlib · Seaborn · Power BI · Streamlit · UI/UX · GitHub · Git · VS Code · Jira
- Built a Graph Neural Network pipeline over state-of-the-art event & tracking data, representing every shot as a heterogeneous graph of all players, the ball, as well as the instantaneous spatial & physical context.
- Trained and benchmarked Heterogeneous Graph Transformers (HGTs) for shot classification and goal-probability (xG) estimation, successfully proving statistically significant gains in shot outcome classification, and their respective goal-probability estimation & calibration over traditional xG models.
- Written recommendation letter from Dr. Gabriel Anzer is available.
- Selected by David Sumpter (and his committee) as 1 of 4 finalists for Twelve Football's “Pitch to the Pros #3”.
TECH STACK: Python · PyTorch · PyTorch Geometric (PyG) · Heterogeneous Graph Transformers (HGTs) · scikit-learn · NumPy · pandas · Matplotlib · Seaborn · football event & tracking data · Git · VS Code