(Senior) Data Scientist, Operations Research (m/f/d)

About Freshflow

Freshflow builds the AI operating system for the fresh food supply chain, starting with ordering and inventory optimisation for fresh departments. Grocery retailers across Germany and France, including EDEKA, REWE, Carrefour, Intermarché and E.Leclerc, use Freshflow every day to cut food waste and keep shelves full. Stores accept 93% of our order proposals, and we recently closed a $10M Series A.

Behind every order proposal is a decision under uncertainty: how much of a perishable item to order when demand is uncertain, shelf life is short, and items come in fixed case packs. Order too much and it ends up as waste; order too little and the shelf is empty. Getting that trade-off right, item by item and store by store, is what our customers pay us for.

The role

You join a small team that owns the whole range of data science at Freshflow, from forecasting to the ordering decisions built on it. You'll be the main owner of operations research and decision-making, and you'll also work on whatever problem matters most at the time, forecasting included.

We judge our work by what it does in stores: less waste, fewer stockouts, more profit. That makes this a pragmatic role. The right approach is the one that delivers the most value to customers, not the most elegant one on paper. Sometimes that's a well-designed optimisation model; sometimes it's a simple rule that ships next week.

Who we're looking for: ideally you bring an OR background and experience applying it to real-world decisions in a business setting. If you don't have deep optimisation experience yet, but you have a strong quantitative background and want to grow into owning this area, we want to hear from you too.

What you'll do

• Own how forecasts become decisions: the ordering policy and inventory optimisation that turn uncertain demand into order proposals stores accept. Problems you'll work on include:
  • Ordering perishables with limited shelf life, fixed case packs and delivery schedules.
  • Balancing the cost of ordering too much (waste) against ordering too little (stockouts), item by item.
  • Deciding when a simple rule beats a complex model, and proving it.
• Work across the full data science stack: forecasting, inventory simulation, backtesting and monitoring, wherever the team needs you.
• Prove impact in stores: design backtests and live evaluations that show the effect on waste, stockouts and profit, and respond quickly when performance drops.
• Own quality end to end: from input data through forecasts and optimisation to live order proposals, with well-tested, reviewed production code. Like everyone on the team, you're a data detective: when an order looks off, you trace it back to the root cause and fix it at the source rather than patching the symptom.
• Build with AI: use agentic coding tools to work fast, and work with our AI engineers on product features such as order proposals that explain themselves to store managers.
• Work with Customer Success and Engineering to turn store-level findings into product improvements.

What you bring

• A background in operations research or another quantitative field, such as industrial engineering, applied mathematics, statistics, physics, computer science or economics, ideally applied to real-world decisions in areas such as inventory, supply chain, pricing, scheduling or logistics. Depth of thinking matters more than the specific discipline.
• Comfort with uncertainty: probability, stochastic models and simulation.
• Strong Python and SQL skills, comfort with large, messy real-world data, and the instinct to dig until you know why a number looks wrong.
• Hands-on experience taking models into production, where real decisions depend on them, and keeping them running reliably.
• Use of AI coding tools in your daily work and eagerness to keep improving your agentic workflows.
• The ability to turn complex analysis into clear decisions for product, Customer Success and store teams.
• Pragmatism: you pick the approach that brings the most value, can explain the trade-offs, and are happy to ship something simple when it does the job.
• Openness to working on any problem the team faces, not only optimisation.
• Fluent English.

Nice to have

• Stochastic inventory theory: newsvendor models, service levels, perishable inventory.
• Optimisation methods such as stochastic or dynamic programming, simulation-based optimisation, or mixed-integer programming with solvers.
• Probabilistic forecasting, such as quantile regression, distributional gradient boosting or conformal prediction.
• GCP (BigQuery, Cloud Run, Vertex AI), dbt, workflow orchestration (Cloud Composer, Airflow), Git, containers and CI/CD.
• Supply chain, demand planning, perishables or grocery retail experience.
• German or French.
• Startup or scale-up experience.

What we offer

• Equity participation (VSOP).
• Work from anywhere in Germany, or hybrid from our Berlin office, with relocation support if you'd like to move to Berlin.
• A real-world problem with measurable impact: less food waste, fewer stockouts, more profit.
• A small, senior team where you own your area from day one.
• A fast hiring process: screening call, case study and interviews, then an offer, typically within 2–3 weeks.

How to apply

Send your CV to jobs@freshflow.ai. Tell us briefly about a time you chose a simpler approach because it delivered more in practice.