Lead Data Scientist - Telematics

We don't build pricing models, we build the driving intelligence that feeds them. Our mission is fairer insurance, priced on how people drive, not on who they are, and safer roads. The Telematics team turns raw phone sensor data into meaningful signals about how people drive. Using signal processing and machine learning on high-frequency GPS, accelerometer, and gyroscope data, we extract the behavioural features that power Zego's understanding of driving quality, context, and risk.

The Telematics Data Science team collaborates daily with engineering, product, and actuarial colleagues across the UK and PT. We're one team across Portugal and the UK. PT and UK roles carry the same scope, ownership, and progression. Decisions get made where the work happens, not in a single headquarters. This isn't a solo remote seat; you'll join an established team of peers.

Few teams use phone sensor data to price commercial motor insurance  a Lead Data Scientist, you'll shape how millions of trips are turned into risk signals. You'll own behavioural features and algorithms from hypothesis to production, working at the intersection of data science and engineering on GPS, inertial, and other sensor data. You'll move between quick heuristics and full ML models depending on what the problem calls for.

You ship code to production, not just notebooks.

• Work with raw, high-frequency sensor data: GPS, accelerometer, gyroscope, s isn't warehouse-tabular data: it's noisy, physical, and where the signal actually lives. You'll be processing more than 250k trips per day.
• Research new behavioural features and detection algorithms: read the literature, try ideas, kill the ones that don't survive contact with real data.
• Design and build behavioural features and factors that feed Zego's understanding of driving quality, context, and risk.
• Take ideas from hypothesis to production: prototyping in notebooks, then writing the production-grade Python and SQL that scales.
• Prioritise simple, robust solutions, rule-based when that's enough, ML when it's warranted.
• Lead experiments, validate impact with data, and automate insight generation.
• Collaborate closely with software engineers, product managers, and and actuaries to get features into the pricing path and measure their effect on real policies.
• Translate complex sensor data into clear findings that non-technical stakeholders can act on., Teams work better with time to collaborate and space to get things done. We call it Zego Hybrid: some of us are in our central London office weekly, others monthly or quarterly. It is about finding the balance between face time and focus that produces great work and a healthy life around it.

We also make a point of getting everyone in the same room. Teams come together every quarter, and once a year the whole company does, properly. It is a serious investment and consistently one of the best parts of the year.

• MSc (or integrated MEng/MSc) in a quantitative field:
  Engineering, Computer Science, Physics, Mathematics, or similar. This is a core role that requires strong engineering foundations.
• Working knowledge of digital signal processing or sensor physics. You can reason about noise, sampling, filtering, and the physical meaning behind a signal, not just its numbers.
• Proven track record delivering data science or data engineering projects into production. You write production-grade code, not just notebooks.
• Strong Python and SQL (we use Snowflake). Comfortable with the scientific Python stack:
  Polars, Pandas, Num Py, Sci Py, and scikit-learn.
• Experience developing and evaluating ML models on tabular data classification or regression tasks where evaluation matters as much as model choice. Think passenger-vs-driver detection, transport-mode classification, or score predictiveness.
• You've designed, built, and maintained data pipelines from scratch: reliable, observable, and scalable.
• Fluent with AI coding assistants as part of daily engineering work. We use Claude Code across the team and expect candidates to be comfortable working with tools of this kind.
• Strong communicator: you translate ambiguous problems into structured, testable ideas, and share insights clearly with technical and non-technical audiences.
• Growth mindset: curious, open to feedback, driven to keep improving.
• You work AI-first. You will use AI daily here, and we mean daily. You do not need to arrive an expert, but you do need to arrive curious, experiment fast, and take ownership of getting good quickly. People who wait to be trained will find this uncomfortable.

• Experience applying signal processing to noisy, high-frequency sensor/time-series data (GPS, accelerometer, IMU) in production.
• Exposure to insurance, mobile data, or behaviour modelling.
• Experience with cloud platforms (AWS), containerisation (Docker, Kubernetes), or data orchestration frameworks.
• Familiarity with our ML and tooling stack: MLflow for experiment tracking, DVC for data versioning, Streamlit for…