Senior Data Engineer

Role Description

A senior data engineer on an eight-person delivery team building a video media planning platform. The role owns the reach and frequency engine, the audience API integration, the optimisation and plan-generation APIs, and the data model that supports these capabilities. Java is the primary delivery language.

The role also covers a second, front-loaded workstream: deriving the configuration values that the scoring methodology runs on. The formulae are fixed and owned by the client, but the input values and configuration parameters required by those formulae still need to be determined and validated. This role works alongside the client’s analysts to determine which signals produce which scores, derive and validate those values, and document how they were arrived at. The client’s analysts then maintain them through the platform’s admin layer.

AdTech background is required at a strong general level, and media-measurement fundamentals are required because of the derivation workstream. Depth in video and TV measurement specifically is an advantage rather than a gate.

About the Project

Client: a global media network, engaged through a delivery partner. End users are media planners at one of the network’s agencies.

Product: a video-only, omnichannel media planning web application. A planner defines a campaign — advertiser, budget, flight dates, target audience, objectives — through a six-step wizard. The platform generates budget-optimised plans across video channels, scores them using two deterministic engines, and exports a client-facing Media Plan (PDF) and an operations Activation Plan (XLSX).

Channels in scope: linear TV, BVOD, SVOD, AVOD/FAST, YouTube, social video, video out-of-home. The market is primarily UK/EMEA.

Scoring: two deterministic engines. A weighted channel quality index built from sub-dimension scores, and a reach and frequency model using exponential saturation against effective spend with cross-channel de-duplication. No machine learning is used.

Delivery: three-month fixed window, sprint cadence, hard go-live date. The team covers delivery management, solution architecture, back-end, data, front-end, DevOps and QA automation.

Key Responsibilities

Engineering (primary)

Implement the Reach and Frequency engine, saturation modelling, effective spend calculation, and cross-channel de-duplication.
Build the audience API integration, including the manual-import fallback path.
Build the optimisation APIs, plan generation, and plan variants.
Design the schema and data model that carry the scoring methodology and its configuration.
Work within the eight-person delivery team, sprint cadence, code review, and integration with front-end and platform workstreams.

Derivation and calibration (front-loaded)

Establish which signals produce which scores, working with the client’s analysts, and derive per-channel sub-dimension scores on a 0-10 scale.
Derive reach parameters, saturation constants and maximum reach ceilings per channel and audience band, by fitting curves to observed reach data.
Build and validate the channel overlap matrices used for cross-channel de-duplication, including symmetry and bounds checks.
Assess source data for coverage, comparability and reliability, and state clearly where a derived value is weak or unobtainable.
Document derivation logic, sources and assumptions to a standard the client’s analysts can review.

Engine validation

Implement effective-dated, versioned reference data so that plans use the parameters active on their build date and historical plans remain reproducible.
Validate engine output against expected results and investigate discrepancies, including the de-duplicated reach defect.
Support the client’s analysts in maintaining their values through the admin layer once derivation is complete.

Collaboration and documentation

Participate in methodology sessions with the client’s analysts.
Document derivation logic, data sources, and assumptions to a standard the client can review.
Identify data gaps and dependency risks early and raise them to the Solution Architect and Delivery Manager.

Required Qualifications

Experience

5+ years in production data engineering, including work where the output was a calculated or measured result, scoring, metrics, indices, or analytical outputs, rather than data movement alone.
Delivery inside a team working to a fixed timeline and defined scope.
Direct working contact with external stakeholders — client teams, partners, or vendors.
English: strong written and spoken, C1+ effectively. The role involves client sessions and written documentation.

Technical Acumen

Java — production application code, Spring Boot services.
SQL — expert level, including analytical queries.
REST API design and integration — OpenAPI contracts, integration against third-party services, fallback handling.
PostgreSQL — schema and data model design, including effective-dated and versioned reference data.
Containerised cloud services — Azure preferred; AWS or GCP acceptable.
Testing — JUnit, integration testing, validation of computational output.

Domain Knowledge — required

AdTech at a strong general level — how media is planned, bought, delivered and measured; the main platform types and their roles; how campaign and audience data moves between them.
Media-measurement fundamentals — reach and frequency, effective frequency, audience universes and denominators, and cross-channel de-duplication. The requirement is the underlying reasoning, not fluency in any particular market’s toolset or currency.
Quantitative concepts used in measurement — metric definitions, weighting, normalisation, index construction.

Judgement & Soft Capabilities

Works within a fixed, externally owned methodology without attempting to redesign it.
Holds a technical position in discussion with client-side specialists, and revises it when shown to be wrong.
Communicates clearly with non-engineering stakeholders.
Documents work so that an external reviewer can follow the reasoning.
Raises blockers, data gaps, and dependency risks early rather than absorbing them.

Nice to Have

Depth in video or TV measurement — GRPs/TVRs, OTS, co-viewing, panel-based measurement, incremental reach.
Digital video signal quality — viewability, completion rates, screen environment, ad clutter.
CTV and programmatic video experience.
Background in an agency, broadcaster, or measurement vendor.
Experience with media planning or buying tooling.
Experience implementing an analytical method defined by someone else.

Ideal Candidate Profile

A senior data engineer from an adtech background whose work has been analytical rather than purely infrastructural – someone who has built pipelines and data models in service of calculation and measurement, not just throughput.

You write production Java code and design data models that hold up under versioning, auditability, and reproducibility requirements – and you are equally at ease with quantitative work when a value has to be derived, a curve fitted, or a result checked. You are not a specialist analyst, but quantitative work is something they seek out rather than tolerate.

You understand how audiences are measured. Reach and frequency, de-duplication and audience denominators are part of their working knowledge rather than terms to look up, because the first weeks of the role are spent establishing what the numbers should be before there is anything to build them into.

You work well inside a defined boundary. The methodology belongs to the client and is not open for redesign. The engineering task is to implement it correctly, calibrate it with evidence, and document how each value was reached.

You are credible in front of the client. You attend sessions with the client’s analysts, explain what they have built and why, ask for the information they need, and flag gaps early rather than working around them silently.