Sr. Software Engineer - Performance - Databricks

P-97

At Databricks, we are passionate about enabling data teams to solve the world's toughest problems. We do this by building and running the world's best data and AI infrastructure platform so our customers can use deep data insights to improve their business. We constantly push the boundaries of data and AI technology, while simultaneously operating with the resilience, security and scale that is critical to making customers successful on our platform. Databricks develops and operates one of the largest scale software platforms; the fleet consists of millions of virtual machines, generating terabytes of logs and processing exabytes of data per day. At our scale, we regularly observe cloud hardware, network, and operating system faults, and our software must gracefully shield our customers from any of the above.

As a performance engineer, you will work closely with multiple teams across the company to evaluate the performance of products and features, identify performance bottlenecks, and partner with engineers to solve performance and scalability issues. This implies, among other teams, setting performance targets for various software releases, guiding teams to develop performance benchmarks, running competitive benchmark analysis for different Databricks products, doing deep dive analysis to identify performance issues and fix them.

The impact you will have:

• Identify performance limitations of the entire stack based on telemetry, customer signals, PoCs, and competitive benchmarks, that will result in the best performing system across the industry, when resolved. Dimensions include latency, data and compute scalability, concurrency, cost, and price to performance ratio. Impact spans all cloud providers and all major areas.
• Set the performance expectations for all cross-cutting efforts early on through specialized benchmarks capturing the intended customer user journeys, and make sure they are met before deployed to customers.
• Understand the performance characteristics of the compute instance types, storage layers, and all cloud services Databricks depends on and deploy optimal solutions to meet the customer demand.
• Work with customers to root cause and mitigate performance problems during production, previews, and POCs.

What We Look For:

• BS (or higher degree) in Computer Science, or a related field
• Experience in the performance analysis discipline. Ability to identify performance issues, root cause problems, and be able to come up with potential solutions. 
• Experience in software development, preferably in large scale distributed systems
• Ability to measure and document the impact of performance features to existing customers, such as possible regressions for certain workloads, their extent, and which customers will be affected.
• Ability to build strong working relationships with developers and field engineers to facilitate triaging and mitigation of performance problems.