- Analyze and understand end-to-end system performance to enhance researcher throughput and velocity.
- Work across the OS, kernel, networking stack, storage subsystems, memory hierarchy, and CPU layers to optimize compute resource utilization.
- Perform deep-dive analysis using Linux performance tools to identify bottlenecks, latency outliers, and non-scalable patterns.
- Read and understand software layers, from application code through language runtimes, system libraries, and kernel code paths, providing suggestions/PRs that optimize critical paths.
- Free up capacity, reduce costs, and scale models by improving computational efficiency through tuning, configuration, and architectural changes.
- Develop systems to improve performance telemetry and observability
- Conduct proof-of-concept evaluations of new performance technologies (new kernel features, alternative system libraries) and contribute to system design decisions.
- Work collaboratively with development teams to catch performance regressions, review performance-sensitive code, and share optimization learnings across the organization
Below is a list of skills and experiences we think are relevant. Even if you don't think you're a perfect match, we still encourage you to apply because we are committed to developing our people.
- Linux internals expertise: kernel subsystems, scheduling, memory management, virtual memory, cgroups, namespaces, and the networking stack.
- Performance tooling: proficiency with perf, eBPF, Ftrace, strace, sysstat, and related tracing/profiling tools. Ability to develop custom tooling when existing tools fall short.
- Systems-level optimization: Experience tuning system and application performance via sysctls, scheduler policies, NUMA topology awareness, memory allocators, network tunables, and storage I/O configuration.
- Hardware-software interface: Deep understanding of CPU microarchitecture, cache hierarchies, memory subsystems, hardware performance counters, and how software behavior maps to hardware reality.
- Versatile coding ability: Comfortable reading and contributing to code across multiple languages (C, C++, Python, Go, Rust, Java) and navigating unfamiliar codebases quickly.
- Observability and telemetry: Experience building or integrating monitoring and observability infrastructure (Prometheus, Grafana, OpenTelemetry).
- Collaborative and communicative: Ability to analyze performance holistically and to effectively communicate findings, trade-offs, and recommendations to development teams and stakeholders.
The salary range for this role is between $195,000 and $225,000. This range is not inclusive of any potential bonus amounts. Factors that may impact the agreed upon salary within the range for a particular candidate include years of experience, level of education obtained, skill set, and other external factors.
PRIVACY STATEMENT: For information on ways PDT may collect, use, and process your personal information, please see PDT’s privacy notices.