Staff Machine Learning Engineer
AI summary of the role
Staff-level ML engineer owning the AI layer of ServiceNow's CPQ platform: building and evolving production agent architectures (LangGraph, custom Harness runtime, MCP, A2A) that let users configure and quote via natural language.
What you’ll do
- Own the Harness, a from-scratch async agent execution runtime with full-duplex streaming, backpressure, and graceful cancellation
- Design multi-agent orchestration with LangGraph/LangChain for transaction editing, conversational configuration, and multi-product quote planning
- Drive MCP and A2A integration strategy, deciding what stays MCP-first and what moves onto the Harness
- Embed with customer- and product-facing teams to adapt agents to real workflows (Forward Deployed Engineering)
What you’ll bring
- 8+ years building production software, with several years shipping LLM-powered/agentic systems
- Deep hands-on expertise with LangGraph and/or LangChain
- Strong MCP understanding, ideally built an MCP server
- Production async Python (FastAPI, asyncio) with WebSockets and streaming
Technologies
LangGraph · LangChain · MCP · A2A · FastAPI · asyncio · WebSockets · PyTorch · scikit-learn · Spring Boot · React · TypeScript
About ServiceNow
Enterprise AI workflow platform connecting people, systems, data, and agents so large organizations can automate IT, employee, customer, security, and industry operations.
Public · 5000+ people
Employer postings · Data from · Sources