Senior Edge AI/ML Inference Engineer

Socket.dev; Austin (TX); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Competitive salary

ACS Equity Package

Health, Dental, Vision Insurance

Paid Time Off

Job summary

Allen Control Systems is seeking a Senior Embedded Software Engineer - Inference AI/ML to own end-to-end deployment of trained models on edge hardware. You will build and optimize the C++ infrastructure for on-device inference, handling pre/post-processing, memory management, and real-time constraints.

You will work with CV/ML, firmware, and product teams to hit performance targets, profiling GPUs/NPUs/DSPs/FPGA and refining models for latency, memory, and power.

Qualifications

• 10+ years in embedded software or systems engineering with ML deployment on edge devices.
• Strong C++ expertise and CUDA knowledge, with hands-on ML framework experience.
• Experience optimizing models for latency, memory, and power on constrained hardware.
• Familiarity with edge platforms (e.g., NVIDIA Jetson) and Linux/RTOS.

Responsibilities

• Own end-to-end ML model deployment on resource-constrained edge hardware.
• Design, implement, and optimize C++ infrastructure for on-device inference.
• Profile latency, memory, and power; tune models to meet constraints.
• Build tooling for packaging, versioning, and OTA updates to devices.
• Collaborate with CV/ML, firmware, and product teams to set targets.

Skills

C++ proficiency

CUDA

PyTorch

Edge inference runtimes

Computer vision

Education

Bachelor’s or Master’s in CS/EE

Tools

TensorRT

ONNX Runtime

TensorFlow Lite

CUDA-enabled GPUs

Job description

Allen Control Systems is seeking a Senior Embedded Software Engineer - Inference AI/ML to own end-to-end deployment of trained models on edge hardware. You will build and optimize the C++ infrastructure for on-device inference, handling pre/post-processing, memory management, and real-time constraints.

You will work with CV/ML, firmware, and product teams to hit performance targets, profiling GPUs/NPUs/DSPs/FPGA and refining models for latency, memory, and power.