AI Staff Engineer| Hybrid: Chicago, IL or NYC, NY

Circana•Chicago, IL

8d•$160,000 - $200,000•Hybrid

About The Position

Circana is building an agentic ecosystem using its consumer behavior data and platform. This Forward Deployed Engineer (FDE) role embeds with teams across operations, client service, and product to translate their expertise into working software, including AI agents, deterministic automations, MCP servers, and integrations. The engineer works in a small pod with a principal engineer and domain experts, builds on Circana’s agent platform and tooling, and owns whether solutions work in production. FDEs move among embedded builds with business teams, shared platform components, and client-facing solutions. The role calls for close collaboration with domain experts and system owners to understand real workflows and exceptions, determine whether work should be handled by a person, automation, or agent, and build and evaluate the selected solution. It involves integrating with enterprise systems, debugging from traces, maintaining evaluation and regression tooling, and improving the development loop through automation.

Requirements

Track record of shipping production software used by real users, including AI or agentic systems.
Strong integration engineering experience with APIs, data pipelines, databases, and systems not personally built.
Deep practical experience with models, including context management, prompting, structured outputs, and tool design.
Strong Python skills and ability to become productive quickly in an unfamiliar codebase.
Fluent daily use of agentic coding tools; Claude Code is preferred.
Comfort working directly with non-technical experts and persistence in understanding how processes actually work.
Experience running workloads on a major cloud platform.
Clear written communication under time pressure.
Demonstrated self-direction through work initiated, built, or fixed without being asked.

Nice To Haves

Prior forward deployed, solutions, or field engineering experience.
Production experience with agentic systems across multiple model families.
Hands-on experience with an agent orchestration framework such as Google ADK, LangGraph, Microsoft Agent Framework, Claude Agent SDK, or similar, on GCP Vertex AI, Azure AI Foundry, or a comparable AI stack.
Working TypeScript skills and solid SQL skills.
Front-end capability with React, TypeScript, CopilotKit, or similar.
Experience with workflow orchestration such as Airflow and large transactional datasets.
Experience with evaluation and observability tooling; contributions to agent frameworks or MCP servers.

Responsibilities

Embed with domain experts and engineers who own underlying systems; learn how work happens, including exceptions, and resolve ambiguity at the source.
Determine whether a task calls for deterministic automation, an agent, or a person, and build the appropriate solution.
Build agents end to end, including tools, MCP servers, context and prompt design, plan construction, tool result shaping, and guardrails.
Build integrations with enterprise systems, including REST services, databases, file-based feeds, and schedulers, including systems with limited documentation or no API.
Build and maintain evaluation sets, graders, and regression suites to measure solutions against ground truth; investigate failures down to the step that caused them.
Debug from traces by reproducing, isolating, and fixing issues, then adding tests to prevent recurrence.
Automate the development loop by turning repeated manual work into commands, hooks, and subagents.
Move decisions forward with working prototypes.
Work across embedded builds with business teams, shared platform components, and client-facing solutions.