Lead Data Engineer

About Better Collective

Better Collective is a global sports media group, reaching more than 400 million monthly visits across a portfolio of leading sports media brands. We build products and experiences that make sports more engaging and accessible for fans around the world. Behind that sits a large and diverse technology and data landscape, spanning markets, brands and business models. We work in a collaborative environment where people are encouraged to challenge how things are done, take ownership and help shape what comes next.

The domain Paid media is where the data has to pay off. We spend real money across dozens of ad networks every day. Which campaigns scale, which creatives get cut, which markets get more budget: those calls come out of the models this team builds. Ad platform spend and delivery, affiliate and operator revenue, CRM data, all stitched into one user-level view of what a click was actually worth. That model exists and it works. It is also the most demanding domain on our data platform - the most sources, the fastest schema churn, and the shortest patience for a table that landed late. We are looking for the engineer who owns it.

Where this role sits

A central platform team sets the how. You set the what. Better Collective runs a hub and spoke data organisation. A central data platform team owns the platform itself - MotherDuck, dbt, Dagster, DLT, the golden paths, the standards, the infrastructure. Paid Media Insights is a domain team, a spoke, that builds on it. The split is deliberate and we hold to it. The platform team decides how data is produced here. You decide what paid media produces, and you own everything we run. So you are not starting a new stack, and you are not going to spend your year on plumbing that a central team should be solving once for everyone. You are, however, expected to be a demanding customer. When a golden path does not fit paid media, we want you to say so and push for it to change, not quietly build around it. You will report to the Senior Head of Paid Media Insights and work across Paid Media, with the Campaign, Commercial and CRM teams, as well as with the central platform team.

What you will own

• Every data job in paid media. All of it. Freshness, cost, reliability, alerting, and the incident at 07:30 before the spend review. If a paid media model is wrong or late, it is yours, not the platform team's. That accountability is the core of this role, not a footnote to it.
• The end-to-end user-level model. Ad platform spend and delivery, affiliate and operator revenue, and CRM, joined into one view the business trusts enough to move budget on.
• Ingestion of paid media sources, built on the platform's framework. Which sources, which schemas, and what happens the next time a network changes its API without telling anyone.
• The domain's models, from staging through to the marts analysts and business teams actually query, plus the semantic meaning that sits on top of them.
• Reverse ETL back into the ad platforms, so online and offline conversion data lands where it changes bidding.
• The handover. Part of your first year is taking over the paid media jobs the central team still runs on our behalf, and moving that ownership to where the context actually lives.

What you will lead

A group of engineers, and the standard they build to. You will guide and mentor a group of data engineers in Paid Media Insights, as their technical lead. That means setting the standard for how this domain builds, reviewing their work, unblocking them, and deliberately growing them: giving away the interesting problems rather than keeping them. You will also shape the domain's roadmap together with the Senior Head of Paid Media Insights, the analysts and the business teams, and you will represent paid media in platform conversations, including how we support machine learning and data science on top of this data. This is a hands-on technical leadership role, not a step away from building. You will be in the code.

What we are looking for

• More than a stack match, we are looking for a particular way of working. You take ownership, and it shows up in the boring parts. Plenty of engineers want to design the model. You also want the pager. You would rather own a domain end to end, unglamorous parts included, than build one piece and hand off the consequences.
• When things get tricky, you follow through. You break complex systems into simple parts. Paid media data is genuinely messy: attribution windows, late conversions, identity across platforms, and dozens of source APIs that all disagree. You can hold that complexity and still hand the business something they can reason about.
• You raise the people around you. You are the person colleagues come to for technical advice, and you enjoy that part of the job. You give feedback that lands, you review generously, and you would rather the whole team understand the pipeline than be the only one who does.
• You work well with a platform team. You have strong opinions about standards and you are still happy to work inside someone else's. You know the difference between a constraint worth fighting and a preference worth dropping, and you would rather help fix a shared path than fork it.
• You are already past prompting. AI is part of how you engineer, not something you occasionally ask questions. You have experimented with agents, tooling or AI-enabled workflows, and you have a view on what a data domain has to look like for them to be useful on top of it: documented, standardised, semantically clear. Paid media is likely where we find out first.
• You are close to the business. You will sit with marketers and analysts who are spending budget on what you build. You want that proximity, and you can talk about data with people who do not think in tables.
• You are technically deep where it counts. Production pipelines in SQL and Python, modelling and orchestration tooling such as dbt and Dagster, cloud data warehousing, and a real feel for performance, cost and reliability at scale. Experience with paid media or attribution data, or with machine learning pipelines, is a bonus rather than a requirement.

Not sure you meet every single requirement? That is okay. Your unique experience might be exactly what we are looking for.

Our stack

What you will build on: MotherDuck / DuckDB · dbt · Dagster · DLT · Python · SQL · AWS · reverse ETL and semantic layer tooling. We do not expect you to have worked with all of these. If you have built production pipelines on Snowflake, BigQuery or Redshift and you are curious about where we are going, that transfers.

What we offer

An international company in growth, and a flat structure. Join a strong team where you can create value and make a real business impact, in a dynamic and creative environment with a flat organisational structure. Freedom with responsibility, ample room to grow, and the chance to take on a broad set of responsibilities. Flexible working hours supporting a healthy work-life balance, private health insurance, and office perks including lunch, snacks, gym access and social events, alongside skilled, passionate and fun colleagues.