MLOps & Agentic Platform Engineer (AI Infrastructure)

Hyphen Connect Limited; Seattle (WA); On-site; USD 100,000 - 130,000; Full time

Job summary

Hyphen Connect Limited in Seattle, USA, is looking for a skilled MLOps & Agentic Platform Engineer. In this role, you will manage model registries and develop continuous training loops while implementing A/B testing infrastructure. The ideal candidate will have a strong background in DevOps/MLOps and be adept at deploying scalable microservices on Kubernetes.

Key responsibilities include building observability dashboards to track token usage and agent reasoning paths. This is a fantastic opportunity for someone looking to enhance their career in AI infrastructure.

Qualifications

• Strong background in DevOps/MLOps.
• Proficiency in deploying scalable microservices.
• Experience in building observability dashboards.

Responsibilities

• Manage model registries, continuous training loops, and A/B testing infrastructure.
• Deploy agents as scalable microservices on Kubernetes.
• Build observability dashboards to track token usage, latency, and agent reasoning paths.

Skills

MLOps

DevOps

Scalable microservices

A/B testing infrastructure

Tools

MLflow

Weights & Biases

LangSmith

Job description

MLOps & Agentic Platform Engineer (AI Infrastructure)

Seattle, USA

We are seeking a skilled MLOps & Agentic Platform Engineer. This role involves managing model registries, developing continuous training loops, and implementing A/B testing infrastructure. The ideal candidate will have a strong DevOps/MLOps background and be adept at deploying scalable microservices and building observability dashboards.

Responsibilities

• Manage model registries, continuous training loops, and A/B testing infrastructure.
• Deploy agents as scalable microservices on Kubernetes.
• Build observability dashboards to track token usage, latency, and agent reasoning paths.

Qualifications

• Experience with MLflow, Weights & Biases, or LangSmith.
• Knowledge of building scalable microservice architectures.