Senior Manager, Agentic AI Engineering

EY; Toledo (OH); Hybrid; USD 144.000 - 359.000; Vollzeit

Zusammenfassung

EY is seeking an experienced engineer to build AI agents that drive meaningful work from intent to verified outcomes. You will sculpt execution loops, tools, and controls that enable agents to operate across complex workflows while remaining understandable, steerable, and safe under failures.

The role sits at the intersection of applied AI, distributed systems, and product engineering. Challenger Core develops reusable tools and guardrails; teams test ideas in real conditions, and scalable

Qualifikationen

• Bachelor’s Degree in a relevant field required.
• 9-12+ years of relevant experience in software or applied ML engineering with ownership of production systems.

Aufgaben

• Choose the right execution model for agent architectures.
• Build dependable tools with clear contracts and validated outcomes.
• Engineer agent behavior across planning, routing, delegation and handoffs.
• Create durable context from instructions, user state and environment feedback.
• Bound agent authority with scoped identities and budgets.
• Evaluate complete trajectories with tests that measure verified outcomes.
• Operate long‑running execution with durable state and safe retries.
• Make behavior observable with end‑to‑end traces and replay.

Kenntnisse

Python

TypeScript

Production systems

Distributed systems

Agent delivery

Model APIs

Async work

State management

Security and controls

Ausbildung

Bachelor’s Degree in a relevant field of study

Jobbeschreibung

EY is seeking an experienced engineer to build AI agents that drive meaningful work from intent to verified outcomes. You will sculpt execution loops, tools, and controls that enable agents to operate across complex workflows while remaining understandable, steerable, and safe under failures.

The role sits at the intersection of applied AI, distributed systems, and product engineering. Challenger Core develops reusable tools and guardrails; teams test ideas in real conditions, and scalable