Agentic Systems Designer
Role details
Job location
Tech stack
Job description
At eBay, we're building intelligent systems that empower people-not replace them. Our agentic AI team partners closely with product, engineering, UX, research, and responsible AI to design autonomous and semi-autonomous experiences that help buyers and sellers achieve more with confidence.
As an Agentic Systems Designer, you will shape LLM-powered agentic systems that go beyond simple prompt-response behavior to support multi-step task execution, tool use, learning from feedback, and collaboration across multiple agents. Sitting at the intersection of AI systems, human-centered design, and product strategy, you will define how our agents reason, plan, and act in real-world scenarios-ensuring they are transparent, trustworthy, and aligned with human goals.
In this role, you will be a key partner and thought leader for cross-functional teams as we redefine how people buy, sell, and interact on eBay.
What You Will Accomplish
- Design agentic AI workflows that reliably reason, plan, and execute complex end-to-end experiences using tools, memory, and real-world feedback to improve outcomes for buyers and sellers.
- Translate customer and business goals into modular agent architectures, finding the right balance between automation and human-in-the-loop oversight for different risk and complexity levels.
- Define behavioral, interaction, and system-level patterns for agentic experiences, including how agents decompose goals into actionable steps and explain their decisions to customers and internal partners.
- Shape strategies for short- and long-term memory so agents maintain context across sessions, learn from previous interactions, and continuously improve performance and user satisfaction over time.
- Establish and iterate mechanisms for reflection, error detection, and self-correction, using qualitative and quantitative insights to identify workflow breakdowns and increase trust, safety, and reliability.
- Contribute to the design of multi-agent systems (such as supervisor-worker and networked collaboration models), collaborating closely with engineering, product, UX, research, and responsible AI to ship usable, safe, and explainable solutions at scale.
Requirements
- Experience designing or shaping AI-powered products, agentic workflows, or AI-enabled systems, ideally including exposure to LLM-based or conversational experiences.
- Strong grounding in human-centered design, UX for intelligent systems, or systems thinking, with the ability to reason about both human interaction patterns and the operational complexity of AI services.
- Familiarity with agentic AI concepts such as reasoning loops, tool use, memory management, reflection, and multi-agent collaboration, and an understanding of when and when not to automate.
- Proven ability to collaborate deeply with engineers, product managers, researchers, and responsible AI partners, while advocating for user needs, safety, trust, and transparency.
- Comfort operating in ambiguity, rapidly learning new technical and design patterns, and helping to shape emerging best practices for agentic systems within a dynamic environment.
- Bonus skills such as experience with prototyping tools (for example Figma, Voiceflow, or similar), working with structured configuration formats (such as YAML/JSON), using SQL or Python for interaction/transcript analysis, running A/B tests, or using LLM observability tools to monitor and improve system behavior.
Benefits & conditions
C$124,000 - C$165,500
Base pay offered may vary depending on multiple individualized factors, including location, skills, and experience. The total compensation package for this position may also include other elements, including a target bonus and restricted stock units (as applicable) in addition to a full range of medical, financial, and/or other benefits (including RRSP eligibility, various paid time off benefits, such as PTO and parental leave). Details of participation in these benefit plans will be provided if an employee receives an offer of employment.