Staff Software Engineer, Data Platform at Orca Security

The role

Orca Security, a cloud security company known for agentless technology, is hiring a Staff Software Engineer in Tel Aviv for its Serving Layer team. That team runs the data infrastructure behind the Orca product's UI and API, so anything a customer looks at, whether an asset, an alert or a query, comes through it, at high concurrency and with millisecond latency targets. It takes in data continuously from across the platform and serves it to the product. This is a senior individual contributor role: you decide how data gets stored, modeled and served, and you are who colleagues turn to when something is slow and the reason is unclear.

The stack

• Serving: SingleStore, self-managed on Kubernetes with an operator, plus Trino for analytics.
• Streaming and ingestion: Kafka and Amazon MSK, Apache Flink, Spark Structured Streaming, change data capture pipelines.
• Lakehouse: Apache Iceberg on S3, Parquet, custom compaction.
• Infrastructure: AWS, EKS, Terraform, Helm, multi-region.
• Languages: Python, Go and plenty of SQL.

What you will do

• Take charge of query performance: execution plans, shard keys and indexes, statistics, join strategy and reducing p99 latency under real load.
• Design physical data models (sharding, partitioning, sort keys, retention) and lakehouse table layouts for fast reads.
• Operate SingleStore on Kubernetes: capacity, node layout, operator upgrades, node refreshes, backup and restore, and limiting the impact of failures.
• Tune ingestion from end to end: Kafka topics and consumers, Flink and Spark parallelism, checkpointing, backpressure and data freshness.
• Build query metrics, slow-query processes and load tests that catch regressions before customers notice.
• Agree schemas and query contracts with product and platform teams, and challenge designs that will not scale.

What you need

• 7 or more years of infrastructure, data or backend engineering, with at least 3 of them spent on the performance of analytical queries or distributed databases.
• Deep knowledge of database internals, including execution plans, data distribution and working out the cause of skew, spills, memory pressure and lock contention.
• Hands-on work with a distributed engine (OLAP, HTAP or MPP) such as ClickHouse, SingleStore, Druid, Pinot, Trino/Presto, Snowflake, Redshift or Vertica.
• Good data modeling judgment for read-heavy serving that is sensitive to latency.
• Kafka in production, along with Flink or Spark Structured Streaming, and tuning them for performance.
• Working knowledge of Iceberg, Delta or Hudi and how partitions, file sizes and compaction affect queries.
• Production Kubernetes skills: operators, StatefulSets, resources, storage and safe upgrades of stateful systems.
• Expert SQL, plus Python or Go.
• Terraform, Helm and AWS operations.
• Real ownership of production systems: capacity planning, being on call and investigating incidents.

Nice to have

• SingleStore itself, especially self-managed on Kubernetes.
• Iceberg at scale: compaction, metadata growth and maintenance jobs.
• Multi-region, multi-tenant SaaS.
• Introducing benchmarks, regression gates and load tests into CI for a team.
• A background in cloud security or CNAPP.