Data Scientist

Edikted is a global Gen Z fashion brand. We design, produce and sell trend-led styles to millions of shoppers, with new collections landing every week. In fast fashion the window between spotting a trend and selling out is short, and every unit we buy is a bet on what customers will want weeks from now. Data decides those bets.

About the role

We are hiring Edikted's first Data Scientist, joining our Predictive Analytics team. Your flagship project is demand forecasting, which you will own end to end: from designing the seasonal indexes and item-level models, through validating them against real outcomes, to running them in production.

As our sole data scientist, you will have real ownership: you choose the modelling approach, build it, validate it, and see it through to production. You will work closely with analysts, data engineers and developers, and your models will directly shape what we buy and how much.

The modelling problems you'll own

Today these decisions run on rule-based logic built by the team. Your job is to replace rules with models that learn from data, and to prove the improvement.

• Item-level forecasting: learn seasonality per category from noisy, stock-constrained history, combine it with price, availability and product attributes, and backtest accuracy by horizon.
• New launches: predict early demand for styles with no history, from product attributes such as image, text and price point, and from how similar past launches behaved.
• Censored demand: estimate what would have sold during stockouts, so that out-of-stock periods do not read as low demand.
• Hierarchy and uncertainty: forecasts that stay consistent from category down to size, with prediction intervals so inventory decisions can weigh stockout risk against overstock.
• Explainability: every forecast comes with its drivers, so analysts can challenge it, trust it, and override it with reason.
• Production: automated pipelines, retraining, monitoring and drift detection, from first version to ongoing improvement.

Forecasting is the first project, not the only one. Beyond it, you will take on further problems across fashion, inventory management and shipping optimization.

Requirements

• 4+ years of hands-on data science experience, including models you built from scratch and brought to production yourself.
• Depth in time-series or demand forecasting.
• Hands-on experience with ML approaches to forecasting, such as gradient boosting or hierarchical models, and strong feature engineering.
• Proven ownership: you have scoped a problem, made the trade-offs, delivered, and maintained the result.
• Strong business thinking: you understand the decisions your model feeds, and you judge it by its effect on them, not only by its error metrics.
• Strong statistical foundations: you know what a backtest can and cannot tell you.
• Strong Python.
• You can explain a modelling decision to a non-technical stakeholder and defend it to a technical one.

Advantages

• Retail, e-commerce or supply chain background.
• Inventory optimization or operations research.
• Experience with cold-start or intermittent demand.
• MSc or PhD in a quantitative field.