Machine Learning Engineer, Agentic Systems

ServiceNow is expanding its Agentic AI program and seeking an experienced engineer to run the runtime infrastructure behind Moveworks' AI agents from Mountain View. This is a hands-on distributed systems role focused on orchestration, execution, and real-time delivery of agent responses, not a traditional ML development position. You will help scale the core platform that enables enterprise agents to plan, act, and interact with users at scale.

Responsibilities

• Orchestration engine: a stateful controller that manages long-running agent sessions, coordinating planning, execution, and user interactions across multiple LLM calls and tool invocations.
• Distributed session management: lease-based ownership using DynamoDB conditional writes, heartbeat protocols, and crash recovery through checkpointing.
• Event-driven message pipeline: SQS FIFO queues for ordered delivery, Kafka consumers for event processing, and streaming via gRPC and Socket.IO in real time.
• Structured concurrency: Python asyncio TaskGroups running multiple concurrent tasks per session (polling, heartbeats, output publishing, orchestrator execution) with fail-fast semantics and graceful cancellation.
• Observability infrastructure: OpenTelemetry instrumentation, distributed trace propagation across async boundaries, and custom span lifecycle management for multi-minute sessions.
• Caching and state layers: Redis and DynamoDB key-value stores with per-organization and per-bot scoping, batch read optimization, and hot-reload configuration.

Requirements

• Experience with distributed systems concepts such as consistency models, idempotency, exactly-once delivery, and distributed locking/leasing.
• Proficiency in concurrent and asynchronous programming, including Python asyncio and Go goroutines, with structured concurrency and cancellation handling.
• Background in event-driven architectures, including experience with message queues (SQS, Kafka), pub/sub patterns, backpressure, and delivery guarantees.
• Solid knowledge of infrastructure databases: DynamoDB (conditional writes, transactions) and Redis (connection pooling, pub/sub).
• Observability discipline: OpenTelemetry, distributed tracing, span context propagation, and Prometheus metrics.
• gRPC and protobuf experience: streaming RPCs, service interface design, and robust error handling patterns.
• 2+ years building production backend or infrastructure systems.
• Strong in Python or Go, with preference for proficiency in both.
• Track record designing and operating systems that handle real traffic at scale.
• Ability to work effectively in the face of ambiguity and novel challenges.

Technologies

• Python, Go
• asyncio, DynamoDB, Redis
• SQS, Kafka
• gRPC, protobuf
• Socket.IO
• OpenTelemetry, Prometheus

About the Company

Moveworks is the Agentic AI Assistant platform that lets employees interact with business systems through natural language to find answers and automate tasks. Serving millions of workers at hundreds of large companies, including many Fortune 500 organizations, Moveworks has been recognized for innovation across cloud, AI, and business technology. In December 2025, Moveworks joined ServiceNow to combine enterprise workflow automation with its reasoning engine and natural language capabilities, building a unified AI platform for every person and workflow. You will join a team at the forefront of AI transformation, backed by ServiceNow’s scale and Moveworks’ innovative approach to agentic intelligence.

Work Personas

We classify work arrangements into flexible, remote, or in-office categories based on the role and location. Eligibility for a given work persona may depend on the distance to the nearest ServiceNow office.

Equal Opportunity Employer

ServiceNow is an equal opportunity employer. All qualified applicants will be considered for employment without regard to race, color, religion, sex, sexual orientation, gender identity or expression, national origin, age, disability, or veteran status. Background checks and other legal requirements apply where permitted by law.

Accommodations

We strive to provide accessible experiences for all candidates. If you need a reasonable accommodation to complete any part of the application process or require an alternative method to apply, please contact [email protected].

Export Control

Some positions may involve controlled technology subject to export control regulations. ServiceNow may need to obtain export approvals where applicable, and employment is contingent upon obtaining any required licenses or approvals.