Data Scientist

Our values: Ambition. Humility. Care.

We're fixing fashion's $300 billion fit problem - before clothes are even made.

71% of returns are fit-related, generating 4 billion pounds of textile waste annually and contributing to <10% of global carbon emissions. We use AI and data science to help fashion brands get the cut right in production, preventing returns and waste at the source rather than just predicting them. Our platform acts as an AI co-pilot for buying and production teams, turning 30 years of fashion expertise into data-backed decisions.

Founded by Phoebe Gormley, we've just closed the UK's largest female-founded pre-seed ($4m) led by Albion and SuperSeed, setting us up to build the team and product that will fix fashion's fit problem.

The role:

Fit Collective's success depends on deeply understanding customer data and putting it to work. Working alongside a cross-functional data and engineering team, you'll use data science, analytics, and AI to unlock the insights that prevent fit issues before garments are made.

Questions you'll help us answer:

• Sizing recommendations: Which products fit true to size, which are too large and which are too small? Does it vary by size? When we show users a message to recommend they size up or down, does it reduce returns? Can we design and run an A/B experiment to confirm that?
• Review and return understanding: What do the hundreds of written reviews for a product tell us? Do they indicate there's a fit issue with that product? Or some other kind of issue?
• Benchmarking: Is the return rate for a particular product good or bad, when compared to similar items? Which items are similar to this one? In what way are they similar?
• Return rate prediction: If this product was made true to size, what would we predict the return rate to be? Which attributes of an item of clothing are useful for predicting return rate?
• Production optimisation: before manufacturing starts on an item, what would we predict the return rate on that item to be? Are there particular aspects of the design that we think will lead to higher returns? What changes can be made to improve that?

Your work will directly impact returns, revenue, and waste for major fashion brands.

Our data stack:

• Orchestration: dbt
• Data Warehouse: Snowflake
• Analytics: Hex
• The rest is up to you! As our first data scientist, you'll shape our approach to ML infrastructure, experimentation platforms, and data science best practices

What We're Looking For:

Must-haves:

• 3+ years data science experience: Ideally in production environments where your work has directly impacted users
• Full-stack data science: You handle the entire pipeline - from finding data in messy sources to building feature stores, training models, validating results, and working with engineering to get them into production. You own the end-to-end delivery
• Strong Python and SQL: You'll need both to work independently across the full stack - from querying source data to putting models into production
• ML fundamentals: Solid grasp of core techniques (regression, classification, clustering) and when to use each. You balance model complexity with interpretability based on the business context
• Statistical rigour: You're appropriately skeptical - of your own models, of patterns in small samples, of causal claims from observational data. You know how to design experiments that isolate real effects from noise and confounding
• Startup DNA: You've either worked at an early-stage company or built and shipped side projects from scratch that people actually use
• Product-minded: You care about the "why" behind analyses and can push back constructively when something doesn't make sense for the business

Nice-to-haves:

• Worked with messy, real-world data where ground truth is ambiguous
• Experience building LLM-powered applications (RAG, agents, etc.)
• Familiarity with dbt, Snowflake, or modern data stack tools
• Understanding of the fashion/retail industry
• Experience designing and building your own experimentation framework
• Familiarity with causal inference techniques like propensity score matching, difference-in-differences, instrumental variables, or synthetic controls

What matters most to us:

• You ship. You balance rigour with speed and aren't precious about perfect models when good models solve the problem. You take full ownership of getting insights into production
• You're product-minded and customer obsessed. You care about the "why" behind analyses and understand that the best model is the one that drives business impact
• You're pragmatic about AI. You use LLMs to improve your productivity and explore new approaches, but you know when traditional methods are more appropriate
• You communicate clearly. You can explain complex statistical concepts to non-technical stakeholders and translate business questions into analytical frameworks

Overview:

• Location: UK-based
• Work Style: Hybrid (~2 days per month in the office in London)
• Employment Type: Full-Time
• Salary: £60-100K/year (depending on experience)
• Equity: via employee share option scheme
• Holiday: 25 days + bank holidays, and a day off on your birthday

Why Fit Collective?

We're not just looking for a Full Stack Engineer; we're looking for someone who wants to make a difference. Join us in creating a friendly, innovative, and high-performing culture where your contributions will have an immediate impact on our customers.

If you want to build products that directly reduce waste in a trillion-pound industry while working with a small, focused team, we can't wait to meet you!