Lead Data Scientist and Machine Learning Engineer with extensive experience delivering applied ML and advanced analytics solutions - from PoC to MVP.
Aktualisiert am 10.06.2026
Profil
Freiberufler / Selbstständiger
Remote-Arbeit
Verfügbar ab: 15.06.2026
Verfügbar zu: 100%
davon vor Ort: 100%
Python
Machine Learning
Google Cloud Platform
Forecasting
Data Scientist
Statistik
agiles Projektmanagement
Azure
Git
MLOps
GenAI
German
native
English
fluent

Einsatzorte

Einsatzorte

Berlin (+20km) Dortmund (+20km)
Deutschland
möglich

Projekte

Projekte

3 months
2025-07 - 2025-09

Hourly Price Forward Curve (HPFC) Analysis

Quantitative Analyst Python Seaborn HPFC modeling ...
Quantitative Analyst
  • Analyzed hourly price forward curves (HPFC) and associated risk premiums as part of a strategic pricing review, evaluating price formation through backtesting against realized spot prices, arbitrage analysis using derived standard products and benchmarking against competitor HPFCs and quotations.
  • Identified that the treatment of renewable energy effects caused systematic overpricing of daytime hours relative to competitor HPFCs and exchange-traded standard products, and specified the corrections needed in the pricing logic.
Python Seaborn HPFC modeling energy price formation backtesting electricity markets Bloomberg GitLab
Energy Provider
Remote
4 months
2025-05 - 2025-08

Energy Demand Forecasting for Contract Pricing and Monthly Trading

Data Scientist & Project Manager Python XGBoost statsmodels ...
Data Scientist & Project Manager
  • Designed and prototyped a demand forecasting framework to support contract pricing and monthly trading decisions under uncertainty, combining autoregressive and tree-based models with external drivers (weather and economic)
  • Performed rigorous backtesting and error decomposition to separate scenario and idiosyncratic effects, reducing monthly forecast deviations by 18% compared to the existing rule-based approach.
Python XGBoost statsmodels time series forecasting ensemble modelling feature engineering backtesting energy demand forecasting GitLab MLflow
Energy Provider
Remote / Berlin
8 months
2024-11 - 2025-06

Legal AI Solution for Digital Car Services

Data Scientist Python LangChain LangGraph ...
Data Scientist
  • Further developed a Legal AI solution in production, integrating an additional data source using a RAG-based architecture and extending the existing assessment logic.
  • Analyzed model outputs from live operations and derived incremental improvements to increase evaluation quality and reduce manual review effort and regulatory risk.
Python LangChain LangGraph RAG LLMs NLP production ML GitLab Azure
Digital Car Services Provider
Remote
8 months
2024-04 - 2024-11

Day-Ahead Electricity Price Forecasting Platform

ML Engineer Python cloud architecture microservices ...
ML Engineer
  • Developed a day-ahead electricity price forecasting platform on GCP, integrating market, weather, and system data into automated forecasting workflows.
  • Deployed as an event-driven microservices architecture (Cloud Run, Pub/Sub, BigQuery) to orchestrate data ingestion, processing, model training and inference; implemented ML lifecycle processes.
Python cloud architecture microservices MLOps electricity price forecasting event-driven systems GCP GitLab MLflow Terraform Cloud Build
energy sector
Remote
9 months
2022-10 - 2023-06

Legal AI for Site Lease Contracts

ML Engineer Python NLP document AI ...
ML Engineer
  • Built an end-to-end document AI pipeline to automatically extract key lease terms (expiration, renewal, termination) from thousands of site contracts.
  • Delivered an operational dashboard to centrally monitor 40,000+ sites, supporting timely contract renewals and reducing manual contract management effort.
Python NLP document AI information extraction pipeline development GitLab AWS Apache Superset
Telco company
Remote
7 months
2021-11 - 2022-05

Improvement of Production Mail-Handling System

Data Scientist Python NLP text classification ...
Data Scientist
  • Conducted root-cause analysis of an underperforming production mail-handling system, isolating a specific model and error pattern responsible for the majority of misclassifications.
  • Constructed a new training dataset from system logs, improved feature engineering, and retrained the model, reducing overall error rates by 50% (to ?2%) on a system processing 200,000+ emails per month.
Python NLP text classification feature engineering error analysis production ML GitLab AWS
Shipping company

Aus- und Weiterbildung

Aus- und Weiterbildung

2014
Technische Universität Berlin
PhD in Theoretical Physics

2010
Universität Hamburg
Diploma in Physics

PROFESSIONAL TRAINING
  • Applied Machine Learning Program

Kompetenzen

Kompetenzen

Top-Skills

Python Machine Learning Google Cloud Platform Forecasting Data Scientist Statistik agiles Projektmanagement Azure Git MLOps GenAI

Produkte / Standards / Erfahrungen / Methoden

Profile
Lead Data Scientist with extensive experience leading data science teams and delivering machine learning and AI solutions from exploration to production with measurable business impact. Track record spanning classical ML, forecasting, and GenAI applications, working closely with engineering teams and business stakeholders. Scientific background in theoretical physics.

SKILLS
  • Programming & Data
    • Python
    • SQL
    • Git
    • Docker
  • Data Science Stack
    • pandas
    • NumPy
    • SciPy
    • scikit-learn
    • TensorFlow/Keras
    • XGBoost
    • LangChain
  • Quantitative Modelling & Machine Learning
    • Time Series & State-Space Models
      • ARIMA/SARIMAX
      • Kalman Filter
    • Machine Learning Methods
      • Regression
      • Tree-Based Models
      • Clustering
      • Ensemble Methods
    • Deep Learning
      • LSTM
      • GRU
      • CNN
      • Autoencoders
    • Uncertainty & Interpretability
      • Conformal Prediction
      • Shapley Values
    • GenAI & LLM
      • RAG
      • Document AI
  • Cloud & Production
    • Cloud-native development
      • AWS
      • Azure
      • GCP
    • Production
      • Deployment
      • Monitoring
      • MLOps

PROFESSIONAL EXPERIENCE

11/2025 - heute
Career break

08/2021 - 10/2025
Eraneos Analytics Germany, Berlin
Lead Data Scientist
  • Led multiple cloud-native data science projects from exploration (PoC) to production across diverse domains. Responsible for planning, implementation, stakeholder alignment and delivery quality across the full project lifecycle.
  • Mentored junior and mid-level data scientists through coaching, code reviews, and methodological guidance, contributing to their development through goal setting, feedback, performance reviews, and promotion input.
  • Active member of the Time Series Lab; contributed to the development of industry-specific ML offerings, including a day-ahead electricity price forecasting platform.
  • Contributed to resource planning and team allocation ? matching skills to projects, supporting recruitment and onboarding.

01/2018 - 09/2020
Deloitte GmbH, Berlin
Senior Consultant
  • Developed statistical risk and valuation models for large international banks, including scenario-based stress testing and sensitivity analysis under regulatory constraints.
  • Built internal analytical tools, including a climate stress-testing framework, an operational risk benchmark model, and a balance sheet optimizer for capital allocation.

01/2017 - 12/2017
Ernst & Young GmbH, Berlin
Senior Consultant
  • Developed and validated statistical models for risk assessment and credit rating processes for financial institutions, primarily in audit and compliance projects.

04/2015 - 08/2016
University of South Florida, Tampa
Postdoctoral Researcher
  • Applied the multi-timescale framework to more complex biophysical models of neuronal dynamics, investigating additional physiological phenomena and pathological states. Published multiple papers in peer-reviewed journals

01/2011 - 08/2014
Technische Universität Berlin
Doctoral Researcher
  • Used stability analysis and numerical simulations to develop a multi-timescale analysis framework for complex systems. Applied this framework to study transitions between normal neuronal activity and stroke-, migraine-, and epilepsy-related states. Published multiple papers in peer-reviewed journals as part of the PhD. Funded by the Bernstein Center for Computational Neuroscience (BCCN) Berlin.

Programmiersprachen

Python
SQL

Branchen

Branchen

  • Energy & Utilities
  • Financial Service
  • Telecommunications
  • Life Sciences

Einsatzorte

Einsatzorte

Berlin (+20km) Dortmund (+20km)
Deutschland
möglich

Projekte

Projekte

3 months
2025-07 - 2025-09

Hourly Price Forward Curve (HPFC) Analysis

Quantitative Analyst Python Seaborn HPFC modeling ...
Quantitative Analyst
  • Analyzed hourly price forward curves (HPFC) and associated risk premiums as part of a strategic pricing review, evaluating price formation through backtesting against realized spot prices, arbitrage analysis using derived standard products and benchmarking against competitor HPFCs and quotations.
  • Identified that the treatment of renewable energy effects caused systematic overpricing of daytime hours relative to competitor HPFCs and exchange-traded standard products, and specified the corrections needed in the pricing logic.
Python Seaborn HPFC modeling energy price formation backtesting electricity markets Bloomberg GitLab
Energy Provider
Remote
4 months
2025-05 - 2025-08

Energy Demand Forecasting for Contract Pricing and Monthly Trading

Data Scientist & Project Manager Python XGBoost statsmodels ...
Data Scientist & Project Manager
  • Designed and prototyped a demand forecasting framework to support contract pricing and monthly trading decisions under uncertainty, combining autoregressive and tree-based models with external drivers (weather and economic)
  • Performed rigorous backtesting and error decomposition to separate scenario and idiosyncratic effects, reducing monthly forecast deviations by 18% compared to the existing rule-based approach.
Python XGBoost statsmodels time series forecasting ensemble modelling feature engineering backtesting energy demand forecasting GitLab MLflow
Energy Provider
Remote / Berlin
8 months
2024-11 - 2025-06

Legal AI Solution for Digital Car Services

Data Scientist Python LangChain LangGraph ...
Data Scientist
  • Further developed a Legal AI solution in production, integrating an additional data source using a RAG-based architecture and extending the existing assessment logic.
  • Analyzed model outputs from live operations and derived incremental improvements to increase evaluation quality and reduce manual review effort and regulatory risk.
Python LangChain LangGraph RAG LLMs NLP production ML GitLab Azure
Digital Car Services Provider
Remote
8 months
2024-04 - 2024-11

Day-Ahead Electricity Price Forecasting Platform

ML Engineer Python cloud architecture microservices ...
ML Engineer
  • Developed a day-ahead electricity price forecasting platform on GCP, integrating market, weather, and system data into automated forecasting workflows.
  • Deployed as an event-driven microservices architecture (Cloud Run, Pub/Sub, BigQuery) to orchestrate data ingestion, processing, model training and inference; implemented ML lifecycle processes.
Python cloud architecture microservices MLOps electricity price forecasting event-driven systems GCP GitLab MLflow Terraform Cloud Build
energy sector
Remote
9 months
2022-10 - 2023-06

Legal AI for Site Lease Contracts

ML Engineer Python NLP document AI ...
ML Engineer
  • Built an end-to-end document AI pipeline to automatically extract key lease terms (expiration, renewal, termination) from thousands of site contracts.
  • Delivered an operational dashboard to centrally monitor 40,000+ sites, supporting timely contract renewals and reducing manual contract management effort.
Python NLP document AI information extraction pipeline development GitLab AWS Apache Superset
Telco company
Remote
7 months
2021-11 - 2022-05

Improvement of Production Mail-Handling System

Data Scientist Python NLP text classification ...
Data Scientist
  • Conducted root-cause analysis of an underperforming production mail-handling system, isolating a specific model and error pattern responsible for the majority of misclassifications.
  • Constructed a new training dataset from system logs, improved feature engineering, and retrained the model, reducing overall error rates by 50% (to ?2%) on a system processing 200,000+ emails per month.
Python NLP text classification feature engineering error analysis production ML GitLab AWS
Shipping company

Aus- und Weiterbildung

Aus- und Weiterbildung

2014
Technische Universität Berlin
PhD in Theoretical Physics

2010
Universität Hamburg
Diploma in Physics

PROFESSIONAL TRAINING
  • Applied Machine Learning Program

Kompetenzen

Kompetenzen

Top-Skills

Python Machine Learning Google Cloud Platform Forecasting Data Scientist Statistik agiles Projektmanagement Azure Git MLOps GenAI

Produkte / Standards / Erfahrungen / Methoden

Profile
Lead Data Scientist with extensive experience leading data science teams and delivering machine learning and AI solutions from exploration to production with measurable business impact. Track record spanning classical ML, forecasting, and GenAI applications, working closely with engineering teams and business stakeholders. Scientific background in theoretical physics.

SKILLS
  • Programming & Data
    • Python
    • SQL
    • Git
    • Docker
  • Data Science Stack
    • pandas
    • NumPy
    • SciPy
    • scikit-learn
    • TensorFlow/Keras
    • XGBoost
    • LangChain
  • Quantitative Modelling & Machine Learning
    • Time Series & State-Space Models
      • ARIMA/SARIMAX
      • Kalman Filter
    • Machine Learning Methods
      • Regression
      • Tree-Based Models
      • Clustering
      • Ensemble Methods
    • Deep Learning
      • LSTM
      • GRU
      • CNN
      • Autoencoders
    • Uncertainty & Interpretability
      • Conformal Prediction
      • Shapley Values
    • GenAI & LLM
      • RAG
      • Document AI
  • Cloud & Production
    • Cloud-native development
      • AWS
      • Azure
      • GCP
    • Production
      • Deployment
      • Monitoring
      • MLOps

PROFESSIONAL EXPERIENCE

11/2025 - heute
Career break

08/2021 - 10/2025
Eraneos Analytics Germany, Berlin
Lead Data Scientist
  • Led multiple cloud-native data science projects from exploration (PoC) to production across diverse domains. Responsible for planning, implementation, stakeholder alignment and delivery quality across the full project lifecycle.
  • Mentored junior and mid-level data scientists through coaching, code reviews, and methodological guidance, contributing to their development through goal setting, feedback, performance reviews, and promotion input.
  • Active member of the Time Series Lab; contributed to the development of industry-specific ML offerings, including a day-ahead electricity price forecasting platform.
  • Contributed to resource planning and team allocation ? matching skills to projects, supporting recruitment and onboarding.

01/2018 - 09/2020
Deloitte GmbH, Berlin
Senior Consultant
  • Developed statistical risk and valuation models for large international banks, including scenario-based stress testing and sensitivity analysis under regulatory constraints.
  • Built internal analytical tools, including a climate stress-testing framework, an operational risk benchmark model, and a balance sheet optimizer for capital allocation.

01/2017 - 12/2017
Ernst & Young GmbH, Berlin
Senior Consultant
  • Developed and validated statistical models for risk assessment and credit rating processes for financial institutions, primarily in audit and compliance projects.

04/2015 - 08/2016
University of South Florida, Tampa
Postdoctoral Researcher
  • Applied the multi-timescale framework to more complex biophysical models of neuronal dynamics, investigating additional physiological phenomena and pathological states. Published multiple papers in peer-reviewed journals

01/2011 - 08/2014
Technische Universität Berlin
Doctoral Researcher
  • Used stability analysis and numerical simulations to develop a multi-timescale analysis framework for complex systems. Applied this framework to study transitions between normal neuronal activity and stroke-, migraine-, and epilepsy-related states. Published multiple papers in peer-reviewed journals as part of the PhD. Funded by the Bernstein Center for Computational Neuroscience (BCCN) Berlin.

Programmiersprachen

Python
SQL

Branchen

Branchen

  • Energy & Utilities
  • Financial Service
  • Telecommunications
  • Life Sciences

Vertrauen Sie auf Randstad

Im Bereich Freelancing
Im Bereich Arbeitnehmerüberlassung / Personalvermittlung

Fragen?

Rufen Sie uns an +49 89 500316-300 oder schreiben Sie uns:

Das Freelancer-Portal

Direktester geht's nicht! Ganz einfach Freelancer finden und direkt Kontakt aufnehmen.