Data Engineer in iGaming: Why a Casino Needs Its Own Data Platform

In a typical SaaS product analytics can lag by a day and nothing terrible happens. In an online casino a day’s delay means a media buyer keeps pouring budget for another day into a source that brings in bonus abusers, and anti-fraud spots a series of suspicious payouts only after the money has gone. That’s why the data platform here isn’t a support function but part of the operational loop — and the person who builds it is worth more than the data engineering market average.

How casino data differs from a regular product

The difference comes down to three things: volume, the number of sources and the cost of a mistake.

• Event granularity. The unit of account isn’t an order or a session but an individual spin: the bet, the result, the balance change, a bonus mechanic firing. An average operator generates tens of thousands of such events per second at the evening peak, and all of them are needed broken down by player, game, provider and currency.
• A dozen external sources, all different. The gaming platform, content aggregators (every slot provider delivers its own reports in its own format on its own schedule), payment providers, crypto processing, affiliate programme trackers, the CRM platform, ad accounts. None of them is obliged to be consistent with the others.
• The cost of a discrepancy. If GGR in the provider’s report and in the operator’s warehouse differ by 0.4%, that’s not “rounding error” — it’s a dispute over money with an external company and a question from the CFO. A consistent number is worth more here than an elegant pipeline.

What a data engineer actually builds

Formally — ETL/ELT pipelines and data marts. In practice — four loops, each with its own internal customer and its own data freshness requirements:

• Marketing and affiliates. Cohorts by source, CPA versus LTV, affiliate payout reporting. Attribution matters here: the same player arrives via a promo code, an affiliate click and retargeting all at once.
• CRM and retention. Segments for mailings and bonus campaigns, churn signals, bonus cost. The mart has to refresh quickly — a win-back campaign launches the same day a player stops logging in.
• Anti-fraud and payments. Features for rules and models: deposit velocity, device linkages, anomalies in betting patterns. The same data stream, but with a latency requirement of minutes, not hours.
• Compliance and the regulator. Regular reporting on gaming activity, data for AML checks, support for the GDPR right to erasure and separate data storage by jurisdiction. The boring part of the job, but it’s what the licence rests on.

The business meaning of these numbers is a profession of its own: how they’re read and used in decisions is covered in the article about data analytics in a casino.

The stack and what a working week looks like

The technology set is predictable: SQL and data modelling, Python, Kafka or another streaming tool, Airflow for orchestration, dbt for transformations, a columnar warehouse — ClickHouse, Snowflake or BigQuery — plus Docker, Kubernetes and CI/CD. ClickHouse shows up in the industry noticeably more often than in the market overall: it’s cheap on large event streams and responds fast, which is exactly what a daily decision cycle demands.

Data engineer or BI analyst: which to choose

These two roles often sit in the same team and get mixed up in job ads. The split is simple: the engineer makes sure the number arrives on time and is the same in every report; the analyst is responsible for what it means. The engineer lives in pipelines and data models, the analyst in dashboards, hypotheses and conversations with product. At the same grade the engineer earns roughly 10–15% more: a middle BI analyst in Malta costs €3.8–5.2K, a data engineer €4.2–5.8K. In return the analyst has a shorter path to product roles, and the engineer to platform and leadership roles.

What it pays

The ranges below are monthly net, from the SpinHire profession database. The first column is Malta and Cyprus, the second Poland, Romania and the Baltics, the third remote and hubs such as Tbilisi, Yerevan and Kyiv.

On top comes an annual bonus, usually 10–15% — more modest than in marketing roles, where the variable part can reach half of total income, but it also doesn’t depend on how an acquisition channel performed. Three things push the range up: experience with streams of tens of thousands of events per second, the ability to measure and cut warehouse costs, and a willingness to own the marts the regulator looks at. Juniors are rarely hired for this role: almost all vacancies start at middle, because nobody wants to teach a person engineering and the domain at the same time.

How to get in from mainstream IT

This is one of the few roles in iGaming where domain experience is barely asked about. People come from data engineering in any industry — fintech, e-commerce, telecoms — and from backend development. A fintech background is especially valued: it has the same requirements for consistency of money figures and for change auditing.

1. The stack matters more than the domain. Volumes and tools should come first on your CV: how many events per day, which warehouse, what you orchestrated with. The specifics of the metrics — GGR, NGR, bonus cost — you pick up on the job within a month or six weeks.
2. Technical English at B1+. The team is almost always distributed, and documentation and correspondence with providers are in English. Less conversational fluency is required here than in client-facing roles.
3. Be ready for the data quality question. At the interview you’ll almost certainly be asked how you catch discrepancies and what you do with late-arriving events. That’s the real exam for the role.

The general mechanics of moving into the industry — from how to read job ads to the salary conversation — are covered in the piece on moving from mainstream IT to iGaming. The role almost always allows fully remote work: a 5/2 schedule, usually with no tie to an office on an island.

Where to grow next

Upwards — Senior Data Engineer, then Data Platform Lead with a team of three to six, then Head of Data and CDO. Sideways — analytics engineering closer to product, the anti-fraud platform or ML infrastructure for churn models and player scoring. Demand for the role is consistently high while candidates with experience of large event streams are scarce: this is a case where the market’s bottleneck works in the candidate’s favour.

The full role description with responsibilities, KPIs and tools is on the page data engineer profession, and open positions are in the SpinHire vacancies.

Read also

CareersBackend developer on an iGaming platform: stack, tasks and salariesCareersProduct manager in iGaming: how the job differsCareersCountry manager in iGaming: the person who owns an entire marketDataWho is hiring in iGaming: top 20 companies and 5,620 jobsCareerAffiliate Manager Interview: Questions, Cases and Test Tasks