Performance Engineer - Benchmarking

About this role

We’re hiring experienced performance engineers to make performance measurement at Supabase rigorous, repeatable, and actionable by product teams. This role owns the benchmarking and profiling systems we use to detect regressions, explain performance changes, and publish credible numbers for customers and the market. You’ll build the tooling and the methodology and make it easy for product teams to use it correctly.

You’d join a new performance team as one of its first members, working alongside a deeply seasoned performance engineer, with the team growing to roughly six this year. There’s no legacy process to inherit — you’ll help define how we measure and communicate performance metrics at Supabase.

What you’ll do

• Build and evolve benchmarking, profiling, and load-testing tooling (cloud, database, and end-to-end).

• Define benchmark suites and reference workloads that reflect real customer behavior (and evolve them as the products change).

• Establish performance baselines and regression gates across products (variance tracking, environment control, and detection of changes).

• Build mechanisms that turn benchmark results into action: triage workflows, owner routing, and clear, reproducible reports.

• Partner closely with database teams (e.g. Multigres, OrioleDB) to identify bottlenecks and land concrete performance improvements.

• Enable the company to publish credible, repeatable performance numbers for customers and the market.

• Help teams self-serve performance testing and make performance a first-class engineering concern.

Problems you might work on

• Build our continuous benchmarking infrastructure so product teams can spot performance regressions early. Ideally, teams can run benchmarks directly on PRs or locally while developing a feature.

• Analyze and eliminate issues with load generators, workloads or benchmark environments to ensure that benchmarks are reliable indicators for actual product performance.

• Build regression triage tooling that pinpoints what changed (diffing, profiles/traces, query-level attribution).

• Help product teams so they can analyze the root cause of performance regressions in benchmark environments themselves.

What we’re looking for

• Deep experience in performance engineering, benchmarking, and profiling of complex distributed systems. Specifically, you have owned benchmarking systems where reproducibility and controlling variance mattered.

• Strong systems intuition (databases, storage, networking, cloud infrastructure).

• Comfortable working across product, infra, and database teams.

• Pragmatic, tooling-oriented mindset with strong communication skills.

• Depth on at least one end of that stack: Linux kernel internals (virtual memory, scheduling, networking, IO) or database internals (query planning, storage engines, buffer and WAL management)

What We Offer

• Fully Remote

• ESOP

• Tech Allowance

• Health Benefits

• Annual Off-Sites

• Flexible Work

• Professional Development

• ~400 team members

• 60+ countries

• 20+ languages spoken

• 540,000+ community members