Edge ML Engineer: Port Speech Models to Diverse Hardware

Madrona Venture Labs; United States; On-site; USD 140,000 - 210,000; Full time

Job summary

Deepgram in the United States is seeking an Applied ML Engineer on the Partner Platform Engineering team to port speech models to edge devices beyond NVIDIA GPUs. You will adapt models to target kernels and runtimes, work with Embedded AI Engineers on custom kernels, and own end-to-end deployment to edge hardware.

This is an applied ML role, not research or cloud-serving, ideal for senior engineers who have shipped models to non-GPU hardware and want to scale across platforms.

Qualifications

• Hands-on experience deploying ML models to edge or non-NVIDIA hardware in production.
• Experience with model quantization and precision tradeoffs.
• Knowledge of edge/inference runtimes and their conversion toolchains.
• Ability to modify models to fit platforms without breaking accuracy.

Responsibilities

• Port Deepgram speech models to non-NVIDIA and edge platforms with minimal modification.
• Own decisions on quantization and precision, operator substitutions, and graph rewrites.
• Validate port on real hardware with accuracy, latency, throughput, and memory benchmarks.
• Build the deployment path for edge targets with packaging, conversion, and versioning.
• Collaborate with Embedded AI Engineers when custom kernels are required.
• Partner with platform and silicon vendors on runtimes and toolchains.
• Provide feedback to Research and Impeller to ease future ports.
• Expand edge production concerns like automated deployment and observability as the team grows.

Skills

Edge deployment

Model quantization

Python

PyTorch

Automation

Tools

ONNX Runtime

TFLite

ExecuTorch

OpenVINO

Qualcomm AI Engine

Vendor NPU SDK

Job description

Deepgram in the United States is seeking an Applied ML Engineer on the Partner Platform Engineering team to port speech models to edge devices beyond NVIDIA GPUs. You will adapt models to target kernels and runtimes, work with Embedded AI Engineers on custom kernels, and own end-to-end deployment to edge hardware.

This is an applied ML role, not research or cloud-serving, ideal for senior engineers who have shipped models to non-GPU hardware and want to scale across platforms.