Applied AI Engineer: Build Reliable Agent Systems

Clay; New York (NY); On-site; USD 180,000 - 240,000; Full time

Job summary

Clay is building AI agents that research, enrich, and take action on behalf of users. You will work on the team responsible for agent products and the platform that supports them, ensuring these systems are robust, steerable, and trustworthy in production.

You’ll collaborate with product, research-adjacent teammates, and other engineers to turn model improvements into measurable gains in task completion and efficiency, while pushing the boundaries of go-to-market workflows.

Qualifications

• Experience building or shipping production systems with LLMs or agents.
• Strong backend fundamentals in APIs, databases, distributed systems.
• Experience with model or agent evaluation: designing evals, measuring regressions, or turning fuzzy quality questions into measurable signals
• A systems-and-outcomes mindset: care about whether the product actually works for users.
• Comfort debugging messy, real-world failures and a bias toward shipping quickly.

Responsibilities

• Design and iterate on agent behavior across real GTM workflows.
• Map manual, multi-step workflows into agent-driven flows.
• Build and run evals that measure whether an agent actually completed the task correctly.
• Analyze real failures in production and systematically improve robustness.
• Work with product to take agent flows from early prototype through beta to GA.

Skills

LLMs/Agents

APIs

Distributed systems

Eval design

Debugging production

Systems-thinking

Prompting

Tool-calling

Tools

React

TypeScript

Python

AWS

Terraform

Datadog

Job description

Clay is building AI agents that research, enrich, and take action on behalf of users. You will work on the team responsible for agent products and the platform that supports them, ensuring these systems are robust, steerable, and trustworthy in production.

You’ll collaborate with product, research-adjacent teammates, and other engineers to turn model improvements into measurable gains in task completion and efficiency, while pushing the boundaries of go-to-market workflows.