Remote AI Agent Engineer: Design & Scale Multistep Workflows

Agileday Oy; United States; Remote; USD 120,000 - 190,000; Full time

Benefits offered by this job

Remote-friendly

Flexible location

Occasional in-person team time

Job summary

Agileday Oy is seeking an engineer to design AI agents that interpret user intent, access context, and use tools to complete multi-step work across a unified business platform. You’ll guide architecture from exploration to production and shape features for real customer workflows.

The role focuses on evaluating reliability, grounding, tool use, latency, and cost, with production-ready design and observability.

Qualifications

• Strong software engineering experience from exploration through production.
• Experience with AI and coding agents to explore ideas, design solutions, write and review code.
• Ability to reason about agent behavior, context, permissions, tool use, and multi-step workflows.
• Sound engineering judgment and interest in evaluating production systems.

Responsibilities

• Design agents that access relevant context, use tools, and carry out multi-step tasks.
• Define how agents work with platform data, permissions, memory, and workflows.
• Build evaluation and observability methods to assess reliability, grounding, tool use, latency, cost, and task completion.
• Design for uncertainty, failures, and recovery in production workflows.
• Use real-world behavior and customer insight to evaluate and improve agent capabilities.
• Contribute to technical direction across product engineering, agent architecture, and reliability.

Skills

Strong software engineering

AI and coding agents

Reason about agent behavior

Production systems judgment

Job description

Agileday Oy is seeking an engineer to design AI agents that interpret user intent, access context, and use tools to complete multi-step work across a unified business platform. You’ll guide architecture from exploration to production and shape features for real customer workflows.

The role focuses on evaluating reliability, grounding, tool use, latency, and cost, with production-ready design and observability.