Edge ML Engineer: SLMs, MLOps & On-Device Deployments

United States Digital Space LLC; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

United States Digital Space LLC is seeking a skilled ML Engineer to fine-tune and train large language models (SLMs) using Hugging Face, TRL, and adapter methods. You will optimize models for inference, deploy to edge devices and local servers, and build end-to-end MLOps pipelines from data ingestion to deployment.

You will monitor model accuracy, latency, and hardware utilization in production, evaluate quality with benchmarking suites, and collaborate across teams to ensure efficient, reliable

Qualifications

• Fine-tune SLMs using Hugging Face and adapters.
• Apply quantization, pruning, and knowledge distillation for lightweight models.
• Deploy models to edge devices, mobile, and local servers with latency targets.
• Build end-to-end MLOps pipelines from data ingestion to deployment.
• Monitor model accuracy, latency, and CPU/GPU usage in production.

Responsibilities

• Fine-tune and train SLMs using Hugging Face, TRL, and adapter methods.
• Optimize models for inference via quantization, pruning, and knowledge distillation.
• Deploy models to edge devices, mobile, and local servers with strict latency targets.
• Build end-to-endMLOpspipelines — from data ingestion to deployment.
• Monitor model accuracy, latency, and hardware utilization in production.
• Evaluate model quality using benchmarking frameworks and custom evaluation suites.

Skills

SLM fine-tuning

Model optimization

Edge deployment

MLOps pipelines

Performance monitoring

Tools

Hugging Face

TRL

LoRA/QLoRA/PEFT

Job description

United States Digital Space LLC is seeking a skilled ML Engineer to fine-tune and train large language models (SLMs) using Hugging Face, TRL, and adapter methods. You will optimize models for inference, deploy to edge devices and local servers, and build end-to-end MLOps pipelines from data ingestion to deployment.

You will monitor model accuracy, latency, and hardware utilization in production, evaluate quality with benchmarking suites, and collaborate across teams to ensure efficient, reliable