Prove It and Make It Felt
Make the Value Legible: Decide how a prediction reaches the customer so they understand it, trust it, and act on it—then report realised impact back to Product and the business in numbers that hold up.
Measure Honestly: Define offline and online evaluation for model quality, drift, and reliability, and design the A/B tests or causal analyses that prove a feature improved customer outcomes.
Deliver
Ship and Operate: Deployment, testing, versioning, monitoring, and drift detection. Delivery is part of the job, not a handoff.
- Embed with Product: Partner with Product Managers and Software Engineers to put ML inside real product workflows—and say clearly when ML isn't the answer.
- Who We Are
- The Owner: You measure your work by whether customers made better decisions, not by whether the model shipped.
Closer to the Data and the Customer: You'd rather spend a week understanding what the data means than a week tuning a model. You know the domain is the hard part.
A Multi-Disciplinary Operator: You'll chase down the data yourself when it isn't ready, and build the pipeline if that's what delivery requires. You prioritise usability, "Time to Insight," and customer trust as much as you do code efficiency.
We know that great talent comes from many backgrounds. If you have shipped a model you are proud of, we want to hear from you!
HOW WE WORK
We build with coding agents. You set direction and targets, review output critically, and build the harnesses—scaffolding, context, tests, review loops—that make the next model faster to ship. The leverage is in the verification: the backtests, eval scaffolding, and data checks that make generated work safe to trust. What You'll Bring Experience: 5+ years in applied data science, ML engineering, or data engineering that included owning models in production—including at least one model you built and shipped into a real product. B2B SaaS experience is a strong plus.
Technical Core: Strong Python and applied ML libraries for tabular problems (scikit-learn, XGBoost or LightGBM, statsmodels or Prophet). Solid SQL expertise is required.
ML & Modeling Depth: Depth in supervised learning, forecasting, ranking, recommendation, or optimisation. You have modelled messy operational data, not benchmark datasets.
Data & Delivery: You build the data you need and ship what you build—dbt and Snowflake modelling, feature pipelines, deployment, monitoring, and drift detection. We're on AWS.
Measurement & Narrative: You've quantified the business impact of a model you shipped—adoption, outcome, ROI—and presented it to people who were never going to read your notebook.
Communication & Collaboration: Excellent communication skills with the ability to explain complex technical trade-offs clearly to product, engineering, and non-technical business stakeholders.
Bonus Points For
Working With Agents: You've used coding agents on real modelling or engineering work, you can tell correct output from merely plausible output, and you invest in the scaffolding that makes the next model faster to ship.
Decision Intelligence: Experience with decision intelligence, forecasting, customer behaviour modelling, workforce/route optimisation, or operational intelligence products.
Leadership: Prior experience as a senior or lead scientist or engineer responsible for guiding technical direction.
LLM / Agentic AI: LLM or agentic workflows shipped into a product.
LOCATION & RIGHT TO WORK
Location: Remote (UK-based) with flexible working options.
Right to Work: Applicants must currently have the legal right to work in the UK. We are unable to offer visa sponsorship now or in the future for this position.
