Drimva

Elastic

Principal Software Engineer - Performance Tuning - Elasticsearch

$160K–$304K

United States +1 · Principal · Full-time · Posted Aug 18, 2026

Location
United States
Canada
Visa sponsorship
Applicants located in or nationals of listed sanctioned countries or regions may require an export license as a condition of employment; nationality does not affect employment eligibility.

About the job

Lead architectural and code-level performance engineering to optimize Elasticsearch’s performance, predictability, and scalability across distributed environments.

What you'll do

  • Lead performance initiatives from design through production, including architectural and code-level optimizations.
  • Profile system behavior, build performance models, and maintain benchmarks and regression detection across stateful and Serverless architectures.
  • Build AI-assisted optimization tools and collaborate across teams to promote performance-first development; mentor engineers.

What they ask for

  • Deep knowledge of Java internals, JVM memory management, concurrency, and writing high-performance, thread-safe, lock-free code.
  • Proven experience profiling and optimizing distributed systems, using benchmarking tools, identifying regressions, and implementing optimizations.
  • Solid understanding of distributed systems architecture, including partition tolerance, cluster state propagation, and scaling.
  • Proven use of AI or advanced tooling to accelerate optimization, debug performance issues, and automate benchmarking workflows.

Nice to have

  • Experience integrating native libraries such as C++, Rust, or SIMD-accelerated code into Java applications.
  • Knowledge of storage engine performance, index modes, or vector search optimization; experience with performance SLAs or search engine internals.

Pay

$159,800 — $252,800 USD (base salary); $191,900 — $303,500 USD (base salary)

Tools

Java, JVM, flamegraphs, JMH, Rally, C++, Rust, SIMD

AI focus

Use AI or advanced tooling to accelerate optimization and debugging, and build AI-assisted harnesses to automate profiling, hypothesis testing, and benchmarking.

Model-written summary · the original posting is authoritative

First seen Oct 27, 2026 · Last verified Nov 5, 2026 (UTC)