Aceolution
The Opportunity
We are seeking an elite Performance Engineer for a short-term consulting engagement to drive critical optimizations in our autonomous vehicle's onboard software stack. This position will be tasked with a deep-dive analysis of our near real-time Linux-based system, identifying and eliminating bottlenecks to ensure maximum efficiency, reliability, and determinism. This is a unique opportunity to apply expertise to one of the most challenging and exciting problems in technology today, with a direct and measurable impact on the performance of our autonomous fleet. What Theyll Do
Conduct comprehensive performance analysis of the entire software stack, from the Linux kernel and drivers up to the application layer Utilize advanced profiling tools (perf, eBPF, ftrace, etc.) to identify and diagnose bottlenecks related to CPU, GPU, memory, I/O, and network subsystems Optimize system and thread scheduling to meet near real-time deadlines and minimize latency in critical processing pipelines Analyze Inter-Process Communication (IPC) mechanisms for opportunities to increase throughput and lower overhead Identify and remedy architectural pitfalls that can lead to performance degradation, memory fragmentation, or unpredictable timing Collaborate closely with core engineering teams to implement recommended changes and architectural improvements Develop and document best practices and performance-aware design patterns for developers to follow Establish key performance indicators (KPIs) and tooling to monitor and prevent performance regressions Who They Are
An expert in systems-level performance engineering with a proven track record of optimizing complex, multi-threaded applications in Linux Deep knowledge of Linux kernel internals, particularly in the scheduler, memory management, and networking stack A master of performance analysis tools and methodologies, capable of creating custom tooling when necessary Strong understanding of modern CPU and GPU architectures and how to write software that leverages them efficiently Proficient in C++ and other systems programming languages Experience squeezing performance out of every layer of the stack, from hardware interrupts to application logic Excellent communication skills, able to articulate complex technical issues and solutions clearly and concisely to both engineers and executives Preferred Qualifications
Prior experience in autonomous vehicles, robotics, aerospace, or high-frequency trading (HFT) industries Experience with real-time Linux patches (e.g. PREEMPT_RT) Familiarity with robotics middleware (e.g. ROS, DDS) and their performance characteristics Knowledge of hardware-specific acceleration (e.g. CUDA, TensorRT)
#J-18808-Ljbffr
We are seeking an elite Performance Engineer for a short-term consulting engagement to drive critical optimizations in our autonomous vehicle's onboard software stack. This position will be tasked with a deep-dive analysis of our near real-time Linux-based system, identifying and eliminating bottlenecks to ensure maximum efficiency, reliability, and determinism. This is a unique opportunity to apply expertise to one of the most challenging and exciting problems in technology today, with a direct and measurable impact on the performance of our autonomous fleet. What Theyll Do
Conduct comprehensive performance analysis of the entire software stack, from the Linux kernel and drivers up to the application layer Utilize advanced profiling tools (perf, eBPF, ftrace, etc.) to identify and diagnose bottlenecks related to CPU, GPU, memory, I/O, and network subsystems Optimize system and thread scheduling to meet near real-time deadlines and minimize latency in critical processing pipelines Analyze Inter-Process Communication (IPC) mechanisms for opportunities to increase throughput and lower overhead Identify and remedy architectural pitfalls that can lead to performance degradation, memory fragmentation, or unpredictable timing Collaborate closely with core engineering teams to implement recommended changes and architectural improvements Develop and document best practices and performance-aware design patterns for developers to follow Establish key performance indicators (KPIs) and tooling to monitor and prevent performance regressions Who They Are
An expert in systems-level performance engineering with a proven track record of optimizing complex, multi-threaded applications in Linux Deep knowledge of Linux kernel internals, particularly in the scheduler, memory management, and networking stack A master of performance analysis tools and methodologies, capable of creating custom tooling when necessary Strong understanding of modern CPU and GPU architectures and how to write software that leverages them efficiently Proficient in C++ and other systems programming languages Experience squeezing performance out of every layer of the stack, from hardware interrupts to application logic Excellent communication skills, able to articulate complex technical issues and solutions clearly and concisely to both engineers and executives Preferred Qualifications
Prior experience in autonomous vehicles, robotics, aerospace, or high-frequency trading (HFT) industries Experience with real-time Linux patches (e.g. PREEMPT_RT) Familiarity with robotics middleware (e.g. ROS, DDS) and their performance characteristics Knowledge of hardware-specific acceleration (e.g. CUDA, TensorRT)
#J-18808-Ljbffr