Lead Analytics Engineer (F/M/X)

We are looking for a Lead Analytics Engineer to be the right hand of our CTO on data, and the right hand of our Global Head of Operations on making a 350-person, four-country operation data-driven.

It is one job made of two halves. If what you want is to go deep on modelling and be left alone, this is not the role. Roughly half of it is sitting with Operations, in the UK, France, Singapore and Hong Kong, arguing about what we should be measuring and why, and then being the person who makes that measurable.

We're not starting from scratch. There's a data lake, dbt on BigQuery, Looker on top, and it works. Software engineers, Data scientist and even some business champions commit to it daily. Nobody owns all of it though, which is where you come in.

This is a hands-on individual contributor role with no direct reports. You will not manage a team, and we are not planning to build one under you. You will set the standard for how data lands in our lake and how it gets used, and hold the rest of the engineering team to it.

You'll join a Product & Engineering team of around 20 people, growing by 50% over the next year.

As Lead Analytics Engineer, you will:

Bring new data into the lake, constantly. This is the steady heartbeat of the role: new sources, new fields from sources we already have, new needs from every part of the business. None of it is rocket science, but our business is a dense one. Four countries, thousands of buildings, an IoT fleet, field engineers, contracts, regulation. Modeling it well is the hard part.

Own the modeling layer. dbt on BigQuery today: models, tests, documentation, reliability. You decide how it evolves, including whether dbt is still the right answer in two years, or when the IoT fleet has grown tenfold.

Define what "good" looks like for our operational KPIs. Our field and operations teams know their business better than anyone. What they don't have is someone technical standing next to them, turning that knowledge into numbers everyone trusts. You'll be that person, with Marine, our Global Head of Operations, as your closest counterpart: deciding what we count, how we count it, and what a result has to show before it means we're winning.

Help one standard become true in four countries. We operate in the UK, France, Singapore and Hong Kong, and good practice invented in one of them should become how the whole group works. Spotting it and rolling it out is Operations' call. Making sure it's made on evidence rather than instinct, and that we can tell afterwards whether it worked, is yours.

Build the analysis the business doesn't have yet: client churn prediction, contract-level economics, engineer productivity, and what our IoT devices tell us about how lifts are actually used.

Keep the platform fast and affordable. BigQuery cost control is your call, not an afterthought.

Extend the platform where it needs extending: UDFs, signed links, LLMs running directly in the warehouse, and whatever else the next problem needs.

Raise the standard around you. Engineers across the team push data into the lake. You set the rules of the road, and you push back when they're not followed.

Use them. Obviously. We'd think it strange if you didn't.

The tooling has moved the goalposts in the last two years, for how we ship and for what we can ship at all. We'd rather talk about what that makes possible than about which model we're on this month. It doesn't remove the need for strong engineers. It makes them more valuable, because someone still has to know what good looks like. Use it however you see fit. You won't be BS'd or FOMO'd about it here.

From the outside the industry looks remote and a bit dull. It's neither, and what you find once you're in it is how tightly coupled everything has to be. Problem detection feeds the device state service, which needs the SLA on the contract, which depends on the customer, and which regulation reads differently in each of our four countries. Very little here is a self-contained feature.

The data you will provide will change how someone's day goes on site. You'll see that for yourself during onboarding. It isn't a day-to-day part of the job, but it's the reason the job matters.

You might be a great fit if you:

Have been the person who owned a data platform end to end, from ingestion through to the numbers a leadership team argues over. Ideally without a large team behind you.

Have done both halves. You've written the pipeline and you've sat in the room where the business decided what to do about the result. Neither one feels like someone else's job to you.

Are fluent in SQL, data lakes, transformation and ETL. We care much less about which tools you learned them in.

Are comfortable being the standard-bearer without the authority. Influence and credibility, not a reporting line.

Can tell an operator they're measuring the wrong thing and be listened to, because you understood their problem first.

Bring structure to messy operational data and are not upset when reality refuses to be clean.

Are at ease working fully remotely across four countries and two continents, and take the initiative to stay close to people.

We do not require experience with our specific stack. If you've never touched dbt, BigQuery or Looker but you've built and owned the equivalent elsewhere, we want to talk to you.

English is the working language of this role.

You've started, run or helped build something of your own, and you judge your work by what it changed rather than the code you wrote.

You've worked in an operations-heavy business: field service, logistics, industrial, energy, anywhere the data describes physical things happening in the world.

You've worked with IoT or time-series data.

You have hands-on experience with dbt, BigQuery or Looker.

You've introduced AI tooling into a data team in a way that stuck.

You've worked across multiple countries and know how differently the same process can look in each one.

We don't expect you to have seen everything already. We do expect you to ramp fast and take ownership quickly.

You'd own the data layer of a company that is currently working out how it wants to run globally. Most of the big questions haven't been answered yet, and you'd be in the room for them.

Senior data people usually end up either close to the warehouse or close to the business. Here it's both, with a direct line to the CTO and to the Global Head of Operations.

And it's full remote, properly. The whole team gets together in person three times a year.

Four stages. We move quickly and we'll keep you posted at every step.

An interview with a member of our team, like Lionel (Principal Engineer) 60 mins 👉 A working conversation about how you've built and owned data platforms, the calls you've made, and what you'd want to own here. Make sure the role is a good fit.

A practical exercise, followed by a 90-minute debrief with David (CTO) and another senior engineer 👉 We send you a problem a few days ahead. Spend 30 minutes on it. No code, no slides, no prep deck. Then we talk it through together. We're interested in how you think, what you ask, and where you push back. The artefact doesn't matter.

A final round: 45 mins with Marine (Global Head of Operations), then 45 mins with 2–3 members of the team 👉 With Marine, a conversation about the business half of the role: how you work with operators, and how you'd approach defining what good looks like. With the team, collaboration style, ownership, and expectations on both sides.