Senior AI Agent Architect for Enterprise Workflows

EY; Columbus (OH); Hybrid; USD 144,000 - 329,000; Full time

Job summary

EY is seeking a Senior AI engineer to build deterministic, observable agent systems that operate across complex workflows. You will craft execution loops, tools, context, and control mechanisms for trustworthy, high-assurance AI agents in production.

This role sits in Challenger Core, EY’s program developing new services and reusable agent patterns. Travel up to 50% and a hybrid office approach are expected as you partner with product, platform, security, and evaluation teams.

Qualifications

• Bachelor’s degree in a relevant field and 9–12+ years of experience.
• Strong track record in software or applied ML engineering with live systems.
• Hands-on evidence of shipping large‑language‑model apps, agents, or complex model‑driven workflows.

Responsibilities

• Choose appropriate execution models for agent architectures with measurable value.
• Build dependable tools and interfaces with validated outcomes and safe recovery paths.
• Engineer agent behavior across planning, routing, delegation, retry, and handoff.
• Create durable context combining instructions, user state, retrieval, and memory.
• Bound agent authority with scoped identities and budgets.
• Evaluate complete trajectories with tests covering success and recovery.
• Operate long-running execution with durable state and safe retries.
• Make behavior observable via end-to-end traces and replay for improvement.
• Protect user control with clear intervention points and safe resume.
• Build shared capability across product, platform, security, and evaluation teams.

Skills

Software engineering

Distributed systems

Python

TypeScript

Agent development

Model APIs

Education

Bachelor’s Degree in a relevant field

Tools

Python

TypeScript

Job description

EY is seeking a Senior AI engineer to build deterministic, observable agent systems that operate across complex workflows. You will craft execution loops, tools, context, and control mechanisms for trustworthy, high-assurance AI agents in production.

This role sits in Challenger Core, EY’s program developing new services and reusable agent patterns. Travel up to 50% and a hybrid office approach are expected as you partner with product, platform, security, and evaluation teams.