Efficient Computer
Performance Library Engineer
About this role
Efficient is looking for Performance Library Engineers to enhance their novel energy-efficient general-purpose processor by developing and optimizing performance libraries. This role involves collaboration across teams to push performance limits while contributing to groundbreaking hardware/software integration.
What you'll do
- Develop and optimize libraries and frameworks for high-performance computing.
- Collaborate with compiler and embedded teams for performance insights.
- Maximize software performance on unique dataflow hardware.
- Engage in hardware/software co-design.
- Write and maintain low-level C/C++ code.
- Utilize AI tools for code optimization and debugging.
What they're looking for
- Experience with RISC, DSP or GPU platforms.
- Software performance analysis.
- Framework and library design.
- CUDA, HIP, or parallel programming models.
- Low-level C/C++ programming.
- Performance profiling and benchmarking.
- Excellent communication skills.
- AI tools for code development.
Benefits
- Competitive salary
- Equity program
- 401K match
- Company-paid benefits
- Paid parental leave
- Professional development opportunities
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Efficient Computer
Efficient Computer builds ultra-low-power processors and the infrastructure needed to develop and optimize them, focusing on energy-efficient computing for AI/ML and general-purpose applications. The company is hiring hardware engineers, software optimization specialists, performance researchers, and infrastructure engineers to advance its processor design and benchmarking platform.
View all jobs at Efficient ComputerLikely interview questions
- Walk us through your experience optimizing software for resource-constrained hardware platforms. Which RISC, DSP, or GPU architectures have you worked with, and what specific performance improvements did you achieve?
- Describe your experience with parallel programming models like CUDA or HIP. How have you approached translating high-level algorithms into hardware-efficient implementations?