Edge AI ML Systems Engineer — In-Memory Computing

TETRAMEM INC; San Jose (CA); On-site; USD 110,000 - 250,000; Full time

Job summary

TetraMem is redefining AI with in-memory computing, delivering efficient edge-to-data-center solutions using RRAM tech. Our team drives innovation, enabling advanced ML for edge devices and data centers alike.

We seek a senior ML/embedded engineer to develop lightweight models, optimize deployment on FPGA/ASIC, and mentor junior engineers while advancing state-of-the-art techniques for power and latency improvements.

Qualifications

• 5+ years of experience or PhD in Computer Science, Electrical Engineering, or related fields.
• Strong experience in machine learning with a focus on edge AI and lightweight model deployment.
• Proficiency in ML frameworks such as PyTorch, TensorFlow, JAX.
• Programming with C/C++, Python and ML model optimization.

Responsibilities

• Develop, optimize, and deploy lightweight ML models for edge AI applications, especially audio.
• Implement and optimize ML models on embedded platforms including FPGA and ASICs.
• Collaborate with hardware and software teams to integrate ML models into production systems.
• Research state-of-the-art ML techniques to improve efficiency, latency and power.
• Improve inference efficiency and model compression (quantization, pruning, distillation).
• Provide technical leadership and mentor junior engineers.
• Publish findings, present at conferences, contribute to open-source.

Skills

Edge AI

Lightweight models

Mentorship

Independent work

Education

PhD or MS in CS/EE

Tools

PyTorch

TensorFlow

JAX

C/C++

Python

Job description

TetraMem is redefining AI with in-memory computing, delivering efficient edge-to-data-center solutions using RRAM tech. Our team drives innovation, enabling advanced ML for edge devices and data centers alike.

We seek a senior ML/embedded engineer to develop lightweight models, optimize deployment on FPGA/ASIC, and mentor junior engineers while advancing state-of-the-art techniques for power and latency improvements.