Yazio - Lead Data Platform Engineer

• You'll feel at home here if: • You've operated a cloud data platform at real scale — event ingestion, streaming, transformation, distributed systems — and you've seen what it takes to grow one 5–10x • You're fluent in production reliability. SLOs, freshness guarantees, replay strategies and silent-failure detection aren't abstract concepts to you • You've designed infrastructure for AI or ML in production — feature pipelines, online/offline stores, low-latency serving — not just used AI in your IDE • You've led engineers before, even a small team. You know what good 1:1s and growth conversations look like, and you'd rather stay technical than become a pure manager • You think in trade-offs out loud — speed vs. cost, simplicity vs. flexibility, build vs. buy — and you take a position when you have enough data, even when it isn't perfect • You see your job as making it easy for other teams to do the right thing. Contracts and tooling, not gatekeeping • You're energised by working at a growth-stage company that's still figuring some things out — and you read that as opportunity, not chaos • B2C consumer products at high traffic, especially mobile-first • Strong experimentation culture • Shipping a user-facing AI feature (not internal tooling) and being able to walk through the infrastructure decisions behind it • OKR ownership at team level • Here's what we currently run in production: • Infrastructure & platform: AWS (EKS, MSK, S3), Kubernetes • Infrastructure & platform: • Event streaming & ingestion: Kafka (AWS MSK), public HTTPS endpoints from mobile, web, backend and payments • Event streaming & ingestion: • Processing & orchestration: Spark, Dagster, dbt • Processing & orchestration: • Warehouse & experimentation: Snowflake, GrowthBook • Warehouse & experimentation: • Observability: Prometheus, Grafana, InfluxDB, Graylog/ELK • Observability: • AI foundations: OpenAI API for product features, Opik for LLM evaluation • AI foundations: • Collaboration: GitHub, structured RFCs, async-first • Collaboration: • If you've worked with Kafka, Spark and Snowflake in some form, you'll be at home. If you came from GCP or self-hosted, the patterns transfer.