Staff Software Engineer - Data Platform Team

Описание

The Serving Layer team at Orca Security owns the data infrastructure for the platform UI and API, handling high concurrency and low latency for all customer assets, alerts, and queries.

Responsibilities

• Own end-to-end query performance: plan analysis, shard-key and index design, statistics, join strategy, and driving down p99 latency under real concurrency
• Design and evolve the physical data models - sharding, partitioning, sort keys, retention - that serve product queries at scale, and the table layouts that make lakehouse reads fast
• Run SingleStore in production on Kubernetes: capacity and node topology, operator-driven upgrades and node refreshes, backup and restore, blast-radius and failure design
• Tune the ingestion path end to end: Kafka topic design and consumer performance, Flink and Spark job parallelism, checkpointing, backpressure, and data freshness
• Build the observability that catches performance regressions before customers do - query-level metrics, slow-query workflows, load-test harnesses
• Partner with product and platform teams on schema and query contracts, and push back on designs that won't hold at scale

Requirements

• 7+ years in backend, data, or infrastructure engineering
• 3+ years focused on distributed database or analytical query performance
• Real depth in database internals: reading and reshaping execution plans, understanding how sharding, partitioning, indexing, and join distribution drive latency, and diagnosing skew, spills, lock contention, and memory pressure in production
• Hands-on experience with a distributed OLAP/HTAP or MPP engine - SingleStore, ClickHouse, Druid, Pinot, Trino/Presto, Snowflake, Redshift, Vertica, or similar
• Strong data modeling judgment for read-heavy, latency-sensitive serving workloads, not only batch or warehouse modeling
• Production experience with distributed streaming - Kafka and at least one of Flink or Spark Structured Streaming - including performance and throughput tuning, not just wiring jobs together
• Working knowledge of a table format (Iceberg, Delta, Hudi) and what partitioning, file sizing, and compaction do to query performance
• Kubernetes fluency in production: operators and StatefulSets, resource management, persistent storage, safely upgrading stateful workloads
• Expert SQL, plus Python or Go
• Infrastructure-as-code and cloud operations (Terraform, Helm, AWS)
• Proven ownership of production systems: on-call, incident investigation, capacity planning

Nice to have

• Direct SingleStore experience, especially self-managed on Kubernetes
• Iceberg at scale: compaction strategy, metadata growth, maintenance jobs
• Multi-region, multi-tenant SaaS
• Experience introducing performance engineering practice to a team - benchmarks, regression gates, load testing as part of CI
• Cloud security or CNAPP domain background

We offer

• High-growth company
• Disruptive innovation in cloud security
• Well-capitalized with a valuation of $1.8 billion
• Respectful and transparent culture
• Opportunity to join the R&D team and contribute to changing the future of cloud security