Own prioritized competitive benchmarking initiatives across search, vector and hybrid retrieval, and analytical and observability workloads.
Build and maintain automated benchmark harnesses, datasets, query sets, and load profiles that reflect real-world usage, including concurrent ingestion and querying.
Design fair comparisons with documented system versions, configurations, tuning decisions, resource budgets, and functional differences.
Evaluate latency distributions, throughput, resource consumption, and cost efficiency; measure retrieval performance at comparable recall or relevance targets.
Adapt and extend workload-appropriate tools such as Elasticsearch Rally, ClickBench, and vector search benchmark frameworks.
Investigate performance differences using profiling, system metrics, and controlled experiments; collaborate with engineering teams to identify causes and validate improvements.
Evaluate relevant self-managed and managed cloud deployments, including serverless offerings, accounting for scaling behavior and service constraints.
Work with engineering and technical content partners to publish reproducible benchmark artifacts and clear technical reports explaining methodology, findings, limitations, and tradeoffs.
Partner with engineering and product management to prioritize benchmark coverage and translate findings into actionable product improvements.
Requirements
Solid software engineering skills with experience building maintainable automation and analysis tools in Python or a relevant programming language.
Hands-on experience evaluating and diagnosing performance in distributed systems, databases, or search infrastructure.
Depth in at least one relevant area (search, vector retrieval, analytical databases, or observability systems) and ability to learn unfamiliar systems.
Experience designing representative datasets and workloads including query patterns, concurrency, and ingestion behavior.
Sound experimental judgment including repeated measurements, variability analysis, warm-up and cache effects, and distinguishing meaningful changes from measurement noise.
Experience building reproducible test environments and documenting configurations, procedures, and results.
Clear technical writing and communication skills with ability to explain complex findings and limitations.
Strong collaboration skills and ability to translate performance evidence into applicable engineering recommendations.
Nice to Haves
Experience with Elasticsearch, Lucene, Rally, ClickBench, or vector search benchmark frameworks.
Experience measuring retrieval quality alongside latency, throughput, and resource usage.
Familiarity with cloud performance and cost analysis including autoscaling and serverless systems.
Public technical writing, open-source contributions, or published benchmark studies.
Perks and Benefits
Competitive pay based on your work, not your previous salary.
Health coverage for you and your family in many locations.
Flexible locations and schedules to craft your own calendar in many roles.
Generous vacation days each year.
Matching donations up to $2000 (or local currency equivalent) for financial donations and service.
Up to 40 hours each year to volunteer for projects you love.
Minimum of 16 weeks parental leave supporting parenthood.