Agentic AI Harness Architect - Moveworks

What You Will Do •  Set the technical vision for the Moveworks agentic AI harness, including planning, tool selection, context management, memory, critique, reflection, adaptation, and recovery. •  Invent and prototype algorithms that help agents complete longer, more complex tasks with less supervision and stronger outcome reliability. •  Design model-agnostic abstractions for combining foundation models, specialized agents, enterprise tools, policies, and deterministic workflows. •  Develop approaches for agent memory and learning from execution traces, user feedback, and task outcomes while preserving enterprise security and privacy. •  Create rigorous benchmarks and evaluation methods for tool-use correctness, plan quality, task completion, groundedness, policy compliance, latency, and cost. •  Explore methods such as iterative planning, ReAct-style execution, self-critique, reflection, test-time computation, multi-agent coordination, and verifier-guided reasoning. •  Build high-fidelity prototypes, identify the ideas that merit investment, and work with platform and product teams to define a practical path to production. •  Partner across Agentic Systems, Search, Relevance, ML Infrastructure, Product, and Security to connect agent intelligence with enterprise knowledge, permissions, and actions. •  Remain hands-on with code and experiments while influencing architecture, mentoring senior engineers, and raising the technical bar across the organization.   What You Bring •  Deep expertise in machine learning or artificial intelligence, paired with strong systems judgment and the ability to reason about end-to-end product architecture. •  A record of meaningful work on AI agents, reasoning and planning, tool use, coding agents, conversational systems, reinforcement learning, or closely related areas. •  Experience turning ambiguous research questions into working prototypes, measurable hypotheses, and durable technical direction. •  Strong understanding of the tradeoffs among model capability, context, memory, latency, cost, reliability, and system complexity. •  Experience designing evaluations for probabilistic systems, including benchmarks that measure task outcomes rather than surface-level response quality alone. •  Excellent programming skills and the ability to work directly in modern ML and agent stacks. Python expertise is expected; experience with production systems languages is valuable. •  Evidence of technical leadership across teams, including the ability to create clarity, influence roadmaps, and guide other senior engineers without relying on formal authority. •  Typically 8 or more years of relevant industry or research experience, or an equivalent record of exceptional technical and research impact. Preferred Experience •  A master's degree or PhD in computer science, machine learning, artificial intelligence, or a related field, or equivalent practical experience. •  Published research, influential open-source work, or widely adopted systems in agents, language models, reasoning, evaluation, or human-agent interaction. •  Experience developing new agent architectures or core AI products in a research-intensive organization. •  Experience with enterprise requirements such as identity, authorization, auditability, privacy, policy enforcement, and safe tool execution. •  Experience moving ideas from prototype to production with distributed systems and infrastructure teams. What Success Looks Like In your first year, you will establish the architecture and research roadmap for the agentic AI harness, deliver prototypes that improve performance on real enterprise tasks, create evaluations that make those gains measurable, and guide successful approaches into the Moveworks Reasoning Engine and broader AI platform.