Senior AI Platform Engineer – Agentic Systems

Moveworks; Mountain View (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Moveworks is building the runtime infrastructure powering its AI agents which orchestrates, executes, and delivers responses to millions of users in real time. You will work at the intersection of large language models and distributed systems to enable agents to plan, reason, and call tools reliably at enterprise scale.

As a Senior Software Engineer, you’ll contribute to agent orchestration, model routing, and state management with emphasis on correctness, observability, and low latency.

Qualifications

• 5+ years building production backend/infrastructure systems.
• Strong in Python or Go (ideally both).
• Experience designing and operating systems that handle real traffic at scale.
• Hands-on experience building production systems that use LLMs or other ML models.

Responsibilities

• Agent orchestration engine: manage long-running agent sessions across LLM calls and tool invocations.
• Model routing and adaptation: route and fine-tune LLMs for enterprise tasks with permissions and data freshness.
• Distributed session management: lease-based ownership and crash recovery.
• Event-driven message pipeline: FIFO queues, real-time streaming via gRPC/Socket.IO.
• Observability infrastructure: OpenTelemetry instrumentation and distributed tracing.
• Caching and state layers: Redis, DynamoDB, batch reads, per-org/per-bot scope.

Skills

Distributed systems

Python

Go

OpenTelemetry

Prometheus

LLMs

NLP

ML models

Tools

DynamoDB

Redis

Job description

Moveworks is building the runtime infrastructure powering its AI agents which orchestrates, executes, and delivers responses to millions of users in real time. You will work at the intersection of large language models and distributed systems to enable agents to plan, reason, and call tools reliably at enterprise scale.

As a Senior Software Engineer, you’ll contribute to agent orchestration, model routing, and state management with emphasis on correctness, observability, and low latency.