Overview We’re seeking a hands-on Performance Test Engineer to design and execute the end-to-end performance strategy for an ad-serving platform (Akka-based Java microservices) targeting
You’ll build the test harness, model real-world traffic, execute large-scale distributed load tests, and turn findings into actionable tuning guidance .
Responsibilities : Own the performance test strategy & plan (load, stress, spike, soak, scalability, failover).
Model traffic for ad-supported streaming (burstiness, fan-out, cache hit / miss, cold-start, geo distribution, p95 / p99 / p999).
Build automated load frameworks & scripts (preferably Locust / Python ; JMeter where appropriate).
Parameterize data, correlations, and think-time.
Orchestrate distributed load generation (cloud workers, containerized runners) to simulate 4–5M concurrent at scale.
Integrate with observability / APM (metrics, logs, traces) to correlate system bottlenecks across app, JVM / GC, Akka dispatchers, network, caches, and databases.
Produce capacity models & SLAs / SLOs dashboards; run performance gates in CI / CD.
Partner with DevOps & developers to recommend tuning (thread pools, connection pools, GC, autoscaling, cache strategies, DB indexes / queries).
Document test design, scenarios, results, and clear remediation plans .
Technology Proficiency Needed : Load tools : Locust (Python) , JMeter ; (nice to have : k6, Gatling).
Scripting & automation : Python (core), Bash; infra spin-up via Terraform / Docker / Kubernetes for load farms.
Metrics / Tracing : CloudWatch, OpenTelemetry, Prometheus / Grafana; log analysis pipelines.
Familiarity with Java service behaviors (Maven / Gradle pipelines, JVM / GC basics); Akka concepts are a plus.
What makes you a great fit 3–5+ years in performance engineering for large-scale, low-latency distributed systems; streaming / ad-tech exposure is a plus.
Demonstrated success hitting strict SLAs (p95 / p99 latency) under millions of users / RPS.
Strong Python and test-automation skills; ability to build maintainable, reusable test frameworks.
Experience designing realistic workload models , synthetic data generation, and distributed load execution in cloud.
Analytical, communicates crisply with stakeholders, converts data into prioritized recommendations .
Logistics Location : Remote (prefer India candidates) Schedule : Must join US morning calls (Eastern Time) as needed.
Start : 1–3 weeks from offer.
Term : Through end of January (likely extension).
Powered by JazzHR
Performance Tester • IN