Senior Applied Research Scientist

About the role

Senior Applied Research Scientist, Agentic Harness

The Agentic Engineering organization at ServiceNow leads a customer-obsessed mission to build a conversational AI that converts enterprise intent into completed work. We advance how enterprise AI reasons, remembers, and executes. The Agent Orchestration team, which you will join, owns the execution core: the agent harness, orchestration runtime, multi-agent coordination, memory management, and evaluation frameworks that ensure robust agent behavior in production. Every autonomous action promised by Otto depends on the capabilities this team delivers.

Joining this team places you at the forefront of AI transformation, supported by ServiceNow's global scale and the agility of a high-growth environment. We seek world-class talent to help extend agentic AI to every employee across every business sector.

As a Senior Applied Research Scientist, you will own critical segments of the agent harness infrastructure that powers enterprise AI execution. This infrastructure layer enables AI agents to reason over real enterprise data, act across workflows, and operate safely at Fortune 500 scale.

Core Responsibilities

You will design and build the orchestration layer that routes inputs, manages context, invokes tools, and handles retries. This harness engineering role ensures reliable data flow and execution continuity across multi-step agentic workflows. You will own the runtime's fault tolerance, latency, and throughput, designing for enterprise workflows that cannot fail silently or non-deterministically.

Observability is a key focus. You will instrument the harness to provide production insights, enabling the team to understand agent behavior and catch failures before customers do. This includes defining where agent logic resides-whether it is a tool call, a sub-agent, a hardcoded path, or a human escalation-and establishing these design standards across the team.

You will build prompt management systems that maintain stability across model updates and configuration changes. These systems support versioning, templating, and systematic evaluation to ensure consistent agent behavior. Evaluation framework ownership falls within this role, including unit evaluations, integration evaluations, and production monitors that measure agent quality, detect regressions, and drive data-informed decisions.

Integration with frontier LLMs completes the core responsibilities. You will manage model routing, fallback strategies, and production cost and latency tradeoffs. Technical leadership is central to this position, involving architecture decisions, code reviews, and mentoring focused on agentic design patterns and production AI discipline.

Team and Impact

The Agentic Engineering organization is customer-obsessed and dedicated to turning enterprise intent into completed work. The Agent Orchestration team you will join is responsible for the execution core: the agent harness, orchestration runtime, multi-agent coordination, memory management, and evaluation frameworks. Every autonomous action Otto promises depends on what this team ships.

By joining this team, you will be at the forefront of AI transformation, backed by the global scale of ServiceNow and the agility of a high-growth environment. We are looking for world-class talent to help extend agentic AI to every employee across every corner of the business.

Qualifications and Expectations

This role requires deep expertise in building and operating agent infrastructure. You must have a strong background in system design for AI execution, with experience in orchestration layers, context management, and tool invocation. Proven ability to ensure reliability, fault tolerance, and performance at scale is essential.

You should be skilled in building observability into complex systems, with practical experience in tracing, cost attribution, and latency visibility. Prompt engineering and prompt lifecycle management experience is required, along with the ability to build evaluation frameworks that measure agent quality and guide improvements.

LLM integration experience is critical, including work with routing, fallback strategies, and cost-latency tradeoffs in production environments. Technical leadership skills are required to elevate engineering standards through architecture decisions, code reviews, and mentoring.

System boundary design is a core part of the role. You will define where agent logic lives and establish design standards that keep agent behavior stable and predictable. This includes decisions on tool usage, sub-agent delegation, coded paths, and escalation paths to human operators.

What You Will Build

You will create the foundational infrastructure that enables enterprise AI agents to operate reliably and safely. Your work will ensure that agents reason effectively over real data, execute workflows without failure, and remain observable and controllable in production. You will establish the practices and standards that allow agentic systems to evolve safely alongside model advancements.

This role offers the opportunity to shape the core infrastructure of agentic AI at global scale. Your contributions will directly impact how enterprises deploy and rely on AI agents to drive real business outcomes. Confirmation of details is available What you'll do

• Meet the bar Practical notes