AI Agent Engineer
$110k - $200k • Toronto • Full-time
Posted 29d ago
About the job
Gigi is seeking an AI Agent Engineer to join their engineering team in Toronto. This role focuses on the end-to-end development of AI agents and workflows for Gigi, an AI media manager for Amazon DSP. While fundamentally a software engineering role involving API design, data modeling, service integration, testing, and debugging, the core focus will be on how Large Language Models (LLMs) reason, utilize tools, maintain context, and take actions for customers, and the software that ensures these actions are reliable. The ideal candidate is a strong software engineer who has begun building with LLMs and agents and is eager to expand their expertise, rather than an AI researcher or someone with extensive LLM experience. As an early employee, you will significantly influence the product, engineering culture, and company direction.
Responsibilities
- Ship agents end-to-end, owning capabilities from data and tools to reasoning, execution, and user interface.
- Build AI into the development workflow using coding agents for planning, implementation, testing, debugging, and review.
- Develop agent judgment, determining autonomous actions versus human escalation and scoring actions by risk and confidence.
- Define the boundary between probabilistic LLM reasoning and deterministic software execution for features.
- Build reliable agentic systems by designing tools, retrieval, memory, orchestration, guardrails, and human approval flows.
- Evaluate and improve agent behavior by creating evaluations and regression tests from customer workflows and production failures.
- Work directly with customers to gather feedback and translate ambiguous problems into shipped product.
Requirements
- A track record of building and shipping production software used by real customers.
- Strong software engineering experience, with depth in backend or full stack development, including system design, APIs, data models, testing, debugging, and operating production software.
- Hands-on experience building with LLMs or agents, demonstrable through work, open source, or substantial personal projects.
- Comfort across the full technology stack, with deep expertise in at least one area.
- Experience using, or a strong desire to become highly effective with, AI coding tools like Codex, Claude Code, or Cursor.
- Curiosity about how LLMs impact software products and development processes.
- Ability to exercise judgment over specifications, ship quickly, and iterate.
- Comfort using AI coding agents to move quickly while maintaining high standards for architecture, correctness, and maintainability.