SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
CyberCube is a leading analytics platform that quantifies digital risk for the insurance industry, trusted by 75% of top European and US cyber insurance carriers and 70% of top reinsurance brokers. The company is backed by Spectrum Equity and operates globally across San Francisco, New York, London, and Tallinn.
You will own the architecture of CyberCube's data platform layer—a system built to serve dashboards, applications, and AI agents with fast, governed, auditable access to sensitive portfolio data. This is a foundational role shaping what "AI-ready data infrastructure" actually means in production.
Key responsibilities:
- Data as a product: Establish contracts, versioning, documentation, SLAs, and self-service onboarding so product teams can independently access the data layer.
- Modelling & migration: Design schemas for transactional and analytical access; lead migration of existing datasets onto new architectures.
- AI-facing interfaces: Build tool definitions, scoped permissions, metadata exposure, and audit trails enabling agents to query data safely.
- Agent vs. pipeline judgment: Determine where agentic automation adds value versus deterministic pipelines; build tooling to support both.
- Platform health: Improve scalability, reliability, and cost across Spark, Iceberg, Airflow, and AWS infrastructure.
- Governance: Own data lineage, quality checks, and isolation for client-sensitive data accessed by humans or agents.
- Technical leadership: Write and review design documents; translate requirements into epics; guide direction across multiple teams.
Tech stack: Python, GraphQL/Hasura, MySQL, Redis, Elasticsearch, Spark, Iceberg, Airflow, Redshift, AWS (S3, Glue, Lambda, RDS, Containers).
What makes this role compelling: You'll set technical direction across several teams, not just manage a single backlog. You'll build production systems (MCP servers, semantic layers, eval harnesses) that ship to real customers, not demos. The work directly impacts how the insurance industry understands and prices cyber risk. AI tools (Claude Code, Cursor, agentic workflows) are embedded in how the team operates daily.
Requirements:
- Strong background in data or backend engineering with production Python and API experience
- Hands-on GraphQL design experience (Hasura is a plus)
- Deep relational database expertise: schema design, indexing, tuning on RDS MySQL/PostgreSQL or equivalent
- Proven experience with Airflow and lakehouse tools (Spark, Iceberg, Redshift, Athena)
- Track record of authoring and reviewing system design documents
- Real experience applying AI to engineering or data work; daily use of AI coding tools
Bonus qualifications: MCP servers or RAG pipelines, agent eval harnesses, AI observability (tokens, latency, failure modes), LangGraph/Claude Agent SDK/OpenAI Agents SDK or custom orchestration, guardrails and prompt-injection defense, Elasticsearch/OpenSearch.
