Remote Edge AI Engineer - On-Device ML Optimization

Bright Vision Technologies; South Windsor (CT); On-site; USD 100,000 - 155,000; Full time

Job summary

Bright Vision Technologies is seeking an Edge AI Engineer to design, optimize, and deploy machine learning models for resource-constrained edge devices, including mobile platforms and embedded systems, with a focus on quantization and hardware-aware optimization.

The role requires deep expertise in model compression, latency and energy efficiency, building cross-platform runtimes (TensorFlow Lite, ONNX Runtime, Core ML), and collaborating with hardware, firmware, and product teams to ship

Qualifications

• Six+ years of ML engineering experience, with edge or mobile AI.
• Strong proficiency in Python and C++.
• Hands-on experience with model compression, quantization, and pruning.
• Experience with at least one major edge inference framework.
• Solid understanding of mobile and embedded hardware architectures.

Responsibilities

• Design and implement edge AI solutions optimized for diverse hardware including mobile SoCs, NPUs, and embedded accelerators.
• Apply quantization, pruning, distillation, and architectural optimization to fit models within edge constraints.
• Tune model performance for latency, energy efficiency, and memory footprint on target hardware.
• Build cross-platform inference runtimes leveraging frameworks such as TensorFlow Lite, ONNX Runtime, and Core ML.
• Optimize models for specific accelerator backends including DSPs, NPUs, and mobile GPUs.
• Implement on-device model update, versioning, and rollback workflows for safe staged rollouts.

Skills

Python

C++

Model compression

Edge framework

Education

Bachelor's or Master's in CS/CE

Tools

TensorFlow Lite

ONNX Runtime

Core ML

Job description

Bright Vision Technologies is seeking an Edge AI Engineer to design, optimize, and deploy machine learning models for resource-constrained edge devices, including mobile platforms and embedded systems, with a focus on quantization and hardware-aware optimization.

The role requires deep expertise in model compression, latency and energy efficiency, building cross-platform runtimes (TensorFlow Lite, ONNX Runtime, Core ML), and collaborating with hardware, firmware, and product teams to ship