Senior Data Platform Engineer

We are looking for a Senior Data Platform Engineer to join the Data Platform at Smartly!

Smartly is building the standard path for governed data sharing across product services, analytics, and AI/ML. The Data Platform team owns the contracts, control mechanisms, publication paths, catalog metadata, lifecycle rules, lineage, permissions, and runtime building blocks that make reusable data safe and practical at scale.

This is a role for someone who wants to shape a platform early, while working on concrete production problems from the start. Smartly already has the production systems, Kafka backbone, PostgreSQL databases, analytical consumers, and AI/ML use cases that need a better data foundation. The challenge is to turn that reality into a standard path that engineers actually want to use.

Our first milestone is focused: centrally managed publishing tables in service-owned PostgreSQL databases, captured through Debezium CDC into Kafka, and registered in DataHub with ownership, lifecycle, lineage, and impact-analysis metadata. Longer-term, the platform expands toward self-service data products, governed materialisations, durable shared state, and reusable data across Smartly’s operational and analytical systems.

This is not a role where you only maintain an existing reporting pipeline. You will help design and build the platform path that other engineering teams use when they publish and consume shared data.

What you will do

• Design and build platform services, tools, and workflows for contract-driven data publishing and consumption.
• Build the first production path for platform-managed PostgreSQL publishing tables, Debezium CDC, Kafka topics, and catalog registration.
• Create GitOps-style workflows where contracts, schemas, ownership, lifecycle, compatibility, policy metadata, and runtime desired state are reviewed, validated, and reconciled as code.
• Build tooling that gives product engineers fast feedback before they publish or change data products.
• Help define how Smartly models data products, publishing contracts, schema evolution, field stability, deletion semantics, access policy, lineage, and cost attribution.
• Operate and improve the CDC runtime model, including Kubernetes workloads, source-specific isolation, offsets, schema history, heartbeats, signal channels, lag visibility, replay, resnapshotting, and recovery.
• Integrate platform metadata into DataHub so engineers can discover data products, understand ownership, trace lineage, and reason about the impact of changes.
• Work with product, infrastructure, data engineering, analytics, and AI/ML teams to make the standard path easier and safer than bespoke point-to-point integrations.

What we are looking for

• Strong experience building and operating production software systems, preferably in platform, infrastructure, backend, or data-intensive environments.
• Deep understanding of distributed systems trade-offs: reliability, idempotency, replay, eventual consistency, ownership boundaries, compatibility, and operational failure modes.
• Practical experience with PostgreSQL or other relational databases, including schema design, migrations, transactions, replication, or operational performance.
• Experience with Kafka, event streaming, CDC, Debezium, or adjacent data-movement technologies. Direct Debezium experience is valuable, but strong fundamentals matter more.
• Experience building software that other engineers use: CLIs, internal platforms, control planes, automation, CI validation, developer tooling, or self-service workflows.
• Familiarity with Kubernetes and cloud or infrastructure platforms, including runtime isolation, deployment, observability, and operational ownership.
• Ability to turn ambiguous architectural direction into practical increments, make trade-offs explicit, and deliver useful production systems without waiting for perfect certainty.
• Clear written and spoken communication in English. This role involves aligning multiple engineering teams around standards, contracts, and ownership boundaries.

Useful experience, but not required

• DataHub or other data catalog, lineage, or governance systems.
• Open table formats such as Apache Iceberg, object-store-backed data lakes, S3, BigQuery, ClickHouse, or analytical storage systems.
• GitHub Actions, policy-as-code, infrastructure-as-code, controllers, operators, or reconciliation-based systems.
• Data contracts, schema registries, Protobuf, Avro, JSON Schema, compatibility checking, field-level metadata, or contract testing.
• Experience supporting analytics, AI/ML, feature-store-style access patterns, or cross-domain product-service integrations.
• Experience with JVM languages and Python, TypeScript, Go, or Ruby.

What we offer you

At Smartly, we offer a place where you can do meaningful technical work while growing with a global team.

• Inclusive global culture built on trust, transparency, open feedback, and respect for different perspectives.
• Global impact through technology used by major brands to plan, create, optimize, and understand digital advertising at scale.
• Wellbeing support, paid holidays, family leave, and sustainable ways of working.
• Equity options, performance-based rewards, competitive compensation, and career development opportunities.
• A flexible hybrid workplace that balances office collaboration with focused remote work. At Smartly we work 3 days a week from the office.

About Smartly

Smartly is an advertising technology company helping brands create, manage, optimize, and understand digital advertising across major media platforms. Our engineering organization builds the systems that connect media, creative, and intelligence capabilities into a product used by customers globally.

We value clear communication, ownership, curiosity, and collaboration across teams. We want Smartly to be a place where people can do strong work, grow, and feel respected.

Not sure you tick every box? Please apply anyway. We care about strong engineering fundamentals, ownership, curiosity, and the ability to build reliable systems with others. Direct experience with every technology listed here is less important than the judgment to learn, reason from first principles, and ship useful platform capabilities.