Job Search and Career Advice Platform

Aktiviere Job-Benachrichtigungen per E-Mail!

Senior Data Scientist: Optimization (f/m/d)

Green Fusion GmbH

Berlin

Hybrid

Vertraulich

Vollzeit

Heute
Sei unter den ersten Bewerbenden

Erstelle in nur wenigen Minuten einen maßgeschneiderten Lebenslauf

Überzeuge Recruiter und verdiene mehr Geld. Mehr erfahren

Zusammenfassung

A pioneering energy technology firm in Berlin is seeking a Senior Data Scientist to lead the architecture of its Energy Management System. You will design optimization frameworks and develop models that address energy flows in real-time. Ideal candidates will have strong backgrounds in mathematical optimization, Python, and statistics. The company offers flexible work hours, ongoing training, and a collaborative environment focused on driving the energy transition.

Leistungen

Flexible working hours
Ongoing training opportunities
Employee benefits
Regular team events

Qualifikationen

  • You are deeply familiar with mathematical optimization and have experience with solvers for MILP, NLP, or MINLP.
  • You have in-depth statistical knowledge and experience in time-series forecasting.
  • You are proficient in Python and can design complex model architectures.

Aufgaben

  • Lead architect of the decision-making logic within the Energy Management System.
  • Define the mathematical core and create high-level optimization frameworks.
  • Mentor junior team members and help navigate complex problem-solving.

Kenntnisse

Mathematical optimization
Statistical knowledge
Proficiency in Python
Time-series forecasting
Understanding of energy systems

Tools

CasADi
Gurobi
Pyomo
Jobbeschreibung

Digitalization and energy transition in one sentence? That’s what we do at Green Fusion!

Our software holistically optimizes energy systems in the real estate sector, helping to combat climate change through digitalization and automation. We reduce emissions and energy consumption, actively advancing the energy transition.

Tasks

As a Senior Data Scientist, you will be the lead architect of the decision‑making logic within our Energy Management System (EMS). You will define the mathematical core of the system: designing how it predicts, reasons, and optimizes energy flows in real‑time, a complex “time‑based decision” problem in the transition to green energy.

Your tasks:

  • Create the high‑level optimization frameworks (MILP, NLP, or Stochastic Programming) to manage residential energy flows across heat pumps, thermal storage, EVs, and batteries.
  • Design and tune closed‑loop control strategies to ensure system stability, robustness against model/reality mismatch, and seamless integration of high‑level optimization with device constraints.
  • Utilize Stochastic & Learning‑Based Control (e.g. Markov Decision Processes (MDPs), Reinforcement Learning, or Model Predictive Control (MPC) to handle the uncertainty of weather, prices, and human behavior.
  • Develop ML models that respect real‑world constraints. You ensure our algorithms “understand” the thermal inertia of a building or the degradation curves of a lithium‑ion battery.
  • Build high‑fidelity simulations to validate algorithm performance against historical data before deploying code to edge devices and cloud environments.
  • You don’t just write formulas; you architect and implement complex models from scratch in Python, ensuring they are robust enough to run in a cloud‑to‑edge environment.
  • Act as a senior voice in technical sessions. You will mentor junior team members and help navigate complex problem‑solving and define the algorithmic requirements that guide our product roadmap.
  • Work closely with Energy Engineers and Backend Developers to translate math into reliable, production‑grade services that save customers money and CO2.
Requirements

We know that nobody fits a job description 100%. If you see yourself in most of these points and are passionate about our mission, we’d love to hear from you!

  • You are deeply familiar with mathematical optimization. You have hands‑on experience with solvers for MILP, NLP, or MINLP (e.g., CasADi, Gurobi, Pyomo).
  • You have in‑depth statistical knowledge and experience in time‑series forecasting, specifically handling uncertainty through stochastic modeling.
  • You are proficient in Python and can design complex model architectures from the ground up, keeping “the big picture” (end‑to‑end thinking) in mind.
  • You enjoy the “messy” reality of hardware. You are eager to learn the specifics of heat storage, hydraulic balancing, and electrical constraints to ensure your code works in the real world.
  • You can explain the “Why” behind a complex stochastic model to a non‑technical stakeholder and lead a brainstorming session on system architecture with ease.
  • Bonus Points: You bring experience in energy usage prediction, Reinforcement Learning, IoT/Edge deployments, or energy management systems (EMS)—a plus, but not a must.
Benefits

🏠 Flexible working hour models, home office, and remote work.

💡 Ongoing training opportunities – whether through job challenges, our open feedback culture, or sponsored training programs, there are always opportunities to learn and grow.

💼 Employee benefits such as Urban Sports Club or Become1.

🌱 Direct impact through your job – with us, you can actively contribute to the energy transition and fight against climate change every day.

💚 We value our team – that’s why regular team events are very important to us.

🙌 The best team that Berlin has to offer – and maybe even beyond. Don’t believe it? Then find out for yourself and apply now!

---
While we may still be considered pioneers today, with you on board we’re ready to take the market by storm – starting in the DACH region, and soon all across Europe.

You’ll join a motivated, open‑minded, and dynamic team passionate about driving the energy transition forward. We believe this mission can only succeed together.

We’re excited to hear from you – Fernanda will be in touch soon!

Hol dir deinen kostenlosen, vertraulichen Lebenslauf-Check.
eine PDF-, DOC-, DOCX-, ODT- oder PAGES-Datei bis zu 5 MB per Drag & Drop ablegen.