Senior Software Engineer, Capacity & Efficiency

About the Team + Role

The Capacity & Efficiency Engineering team builds the software that manages, governs, and reduces Robinhood's AWS cloud spend across compute, databases, and storage, operating at the intersection of cloud infrastructure, data engineering, and FinOps. We own the full lifecycle of cloud cost, from attribution and forecasting to automation that rightsizes infrastructure at fleet scale and keeps a bursty, latency-sensitive trading platform both reliable and cost-effective.

As a Senior Software Engineer on the team, you will own capacity and efficiency for Robinhood's storage and database fleet, from Aurora clusters to S3 and the caches and volumes in between. You will build systems that size this fleet to demand: Aurora reader counts and writer sizes set from market hours, declared events, and live load instead of fixed schedules; databases, caches, and volumes rightsized to actual usage; and S3 data kept in the right storage class with lifecycle rules. You will take recommendations to safe automation applied in production.

This role is based in our Bellevue, WA office, with in-person attendance expected at least 3 days per week.

What You’ll Do

• Build forecast-driven capacity planning for the storage and database fleet, turning traffic history, load tests, and known events (market open, earnings, IPOs, promotions) into plans that scale up ahead and back down after, with an expiry on every temporary capacity hold.
• Design and build a capacity planner for Aurora that sets reader counts and writer instance sizes from demand and load signals (IOPS, CPU, connections, replica lag), replacing fixed scaling schedules.
• Rightsize databases, caches, and block storage (RDS, ElastiCache, DynamoDB, EBS) on instance size, provisioned IOPS and throughput, and idle capacity, keeping them right-sized as load changes.
• Own S3 efficiency at scale: storage class and Intelligent-Tiering strategy, lifecycle and retention policies, noncurrent versions and incomplete uploads, and request costs, using Storage Lens and inventory data.
• Take every recommendation to closed-loop automation with guardrails, progressive rollout, and rollback, handling constraints such as AWS capacity shortages, fallback instance types, and connection pooling.
• Measure outcomes in cost and reliability (savings against baseline, forecast error, utilization, capacity left behind after events) and share forecasts with the pod rightsizing system.

What You Bring

• Deep hands-on experience operating AWS storage and databases at scale, especially Aurora or RDS (MySQL or PostgreSQL) and S3, including replication, read scaling, storage classes, and lifecycle management.
• Capacity planning and forecasting experience: modeling time-series demand, load testing, and translating traffic into instance counts, sizes, and storage growth.
• A track record of rightsizing infrastructure from utilization data (CPU, memory, IOPS, throughput, bytes stored) without hurting latency or availability.
• Strong Python or Go engineering, with experience building automation that safely modifies production infrastructure (guardrails, progressive rollout, and rollback built in).
• Familiarity with AWS storage and database pricing (S3 storage classes and requests, Aurora I/O-Optimized vs. Standard, Reserved Instances) is a strong plus.
• Experience with bursty, event-driven workloads (trading, payments, ticketing) or with Spark, Airflow, or Kubernetes is a plus.