Applied AI Engineer

Mizo; United States; Remote; USD 77,000 - 119,000; Full time

Job summary

Mizo in Montreal or remote is seeking an engineer to design, launch, and operate AI agents that automate service desk tasks for managed IT service providers, moving prototypes into production and improving reliability in real customer workflows.

You will tune prompts, manage agent loops, analyze traces, debug failures, and measure performance, collaborating with a small founding team in a fast-growing environment.

Qualifications

• The role emphasizes practical LLM engineering and production systems experience, including prompt tuning, agent-loop tuning, trace analysis, debugging and evaluation.
• Interest in making agents work reliably with real customer tasks beyond demonstrations.

Responsibilities

• Design, launch, and operate AI agents that automate service desk work for managed IT service providers.
• Bring systems from prototype into production.
• Improve agent behavior and reliability for ticket triage, customer communication, scheduling, resolution, and escalation.
• Inspect production traces and debug agent and tool-call failures.
• Strengthen evaluation practices to measure agent performance and guide improvements.
• Use real production outcomes to inform the next round of system changes.

Job description

Role overview

Design, launch, and operate AI agents that automate service desk work for managed IT service providers. The role carries systems from prototype into production and focuses on making agents dependable in real customer workflows, including how they use tools, respond to failures, and recover when tasks go off track.

Responsibilities

• Build multi-step agent systems with tool use, planning, memory, and autonomous decision-making.
• Take agents from early prototypes through production operation.
• Improve agent behavior and reliability for work such as ticket triage, customer communication, scheduling, resolution, and escalation.
• Inspect production traces and debug agent and tool-call failures.
• Strengthen evaluation practices to measure agent performance and guide improvements.
• Use real production outcomes to inform the next round of system changes.

Requirements

• The available role details emphasize practical LLM engineering and production systems experience, including prompt and agent-loop tuning, trace analysis, debugging, and evaluation; a specific years-of-experience threshold was not provided.
• Interest in making agents work reliably with real customer tasks, beyond demonstrations.

Nice to have

• Experience with autonomous workflows and agentic AI systems.

Benefits and work setup

• Full-time role, based in Montreal or remote.
• Opportunity to collaborate with a small team in a Montreal office and work directly with the founding team.
• The source describes a fast-growing environment with frequent feedback from production use.