Senior AI Platform Engineer: LLM Runtime & Orchestration

Micasa Global; New York (NY); On-site; USD 170,000 - 230,000; Full time

Job summary

Micasa Global is seeking a Principal Engineer / Forward Deployed Engineer to architect and build the execution layer powering enterprise AI Agent platforms. You will own LLM runtime services, agent orchestration, distributed systems, model routing, tool execution, and production AI infrastructure.

The role emphasizes hands-on development of long-running, asynchronous workloads, scalable APIs, and secure cloud-native AI services at enterprise scale.

Qualifications

• Strong software engineering and distributed systems background.
• Production experience with LLMs, AI Agents, Agentic AI, or function calling/workflow orchestration.
• Experience with asynchronous systems, queues, state management, retries and failure recovery.
• Experience with model APIs, structured outputs, prompt/context management.
• Experience with LLM/AI performance optimization.
• Kubernetes, Docker/Containers; cloud-native architecture.
• Experience with Knowledge Graphs.
• Strong production engineering and troubleshooting skills.

Responsibilities

• Design and build scalable runtime services for long-running, asynchronous AI workloads.
• Build AI Agent / Agentic AI execution frameworks.
• Develop LLM model routing and abstraction layers across multiple providers.
• Design execution state management, retries, checkpointing and failure recovery.
• Build reusable APIs and SDKs for AI workflows.
• Develop tracing, telemetry, observability and evaluation capabilities.
• Work with Knowledge Graphs and enterprise knowledge systems.
• Collaborate with Security, Identity and Platform Engineering teams.
• Build secure, reliable, cloud-native AI services at enterprise scale.

Skills

Distributed systems

LLMs / AI Agents

Asynchronous architectures

Model APIs

LLM performance optimization

Kubernetes

Docker / Containers

Cloud-native

Knowledge Graphs

Production engineering

Tools

Kubernetes

Docker

Job description

Micasa Global is seeking a Principal Engineer / Forward Deployed Engineer to architect and build the execution layer powering enterprise AI Agent platforms. You will own LLM runtime services, agent orchestration, distributed systems, model routing, tool execution, and production AI infrastructure.

The role emphasizes hands-on development of long-running, asynchronous workloads, scalable APIs, and secure cloud-native AI services at enterprise scale.