Compiler & Runtime Engineer for AI Accelerator

TetraMem - Accelerate The World; San Jose (CA); On-site; USD 135,000 - 165,000; Full time

Job summary

TetraMem in San Jose, CA, seeks a software engineer to develop, maintain, and optimize the compiler, runtime, and SDK toolchain for analog CIM hardware. You will translate ML models into efficient workloads for the AI accelerator and implement features, tools, and optimizations to boost latency, throughput, memory use, and power efficiency.

Collaborate with compiler, ML, hardware, and validation teams to resolve issues and improve performance.

Qualifications

• Bachelor’s or Master’s degree in CS/CE/EE or related field.
• Experience developing embedded software for microcontroller-based or embedded computing platforms.
• Strong programming in C/C++ for embedded systems and Python for ML development.
• Solid understanding of software engineering principles, data structures, algorithms, and debugging techniques.

Responsibilities

• Develop, maintain, and optimize software components within TetraMem’s compiler, runtime, and SDK toolchain for analog CIM hardware.
• Translate ML models into efficient executable workloads for TetraMem’s AI accelerator architecture.
• Implement features, debugging tools, and low-level optimizations to improve latency, throughput, memory utilization, and power efficiency.
• Support integration of ML frameworks, model conversion pipelines, and deployment workflows for customer applications.
• Collaborate with compiler, ML, hardware, and validation teams to resolve software issues and improve system performance.
• Assist in benchmarking, profiling, and analyzing AI workloads to identify bottlenecks and optimization opportunities.
• Support hardware bring-up, software validation, and demonstration of AI applications on TetraMem compute-in-memory platforms.
• Write clean, maintainable, and well-documented production-quality code following software engineering best practices.
• Participate in design reviews, code reviews, testing, and continuous improvement of the compiler and runtime software stack.
• Stay current with advances in compiler tech, AI frameworks, and computer architecture to improve TetraMem’s software platform.

Skills

C/C++ for embedded systems

Python for ML development

Software debugging techniques

Linux development

CI/CD pipelines

Rust

Education

Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or related field

Tools

Zephyr RTOS

FreeRTOS

Git

Linux

Job description

TetraMem in San Jose, CA, seeks a software engineer to develop, maintain, and optimize the compiler, runtime, and SDK toolchain for analog CIM hardware. You will translate ML models into efficient workloads for the AI accelerator and implement features, tools, and optimizations to boost latency, throughput, memory use, and power efficiency.

Collaborate with compiler, ML, hardware, and validation teams to resolve issues and improve performance.