Hybrid AI Agent Engineer: Build Reliable Workflows

Trace; Montreal (administrative region); Hybrid; CAD 90,000 - 130,000; Full time

Job summary

Trace in Montreal, QC is building AI agents that handle freight operations workflows. We’re seeking engineers to take an agent from idea to a dependable system, collaborating with Noam, our founding engineer, and Lucas, our CTO. You’ll shape both customer-facing and internal agents across tools and APIs.

You’ll fit here if you’ve built agents that use tools for multistep tasks, can debug the software powering them, and explain how you manage context and recover from failures.

Qualifications

• Built agents that use tools to perform multistep tasks and demonstrate behavior.
• Able to debug and modify the underlying software powering agents.
• Explain how you manage context, evaluate behavior and recover from failures.

Responsibilities

• Build customer-facing and internal agents that complete workflows across tools and systems.
• Design agent harnesses: tool interfaces, context management, memory, orchestration, and execution controls.
• Connect agents to APIs and business applications for action.
• Build evaluations, regression tests, and realistic test environments; inspect execution traces.
• Handle ambiguous instructions, failed tool calls, retries, permissions, and escalation.
• Improve reliability, response time, and cost as agents scale up.

Skills

Agent development

Debugging

Context management

Independent

Tools

Claude Code

Codex

APIs

Job description

Trace in Montreal, QC is building AI agents that handle freight operations workflows. We’re seeking engineers to take an agent from idea to a dependable system, collaborating with Noam, our founding engineer, and Lucas, our CTO. You’ll shape both customer-facing and internal agents across tools and APIs.

You’ll fit here if you’ve built agents that use tools for multistep tasks, can debug the software powering them, and explain how you manage context and recover from failures.