Amazon
Mid
SDE II building autonomous AI agent systems that automate Amazon Leo's global ecommerce launch pipeline
A mid-level (SDE II) role on Amazon's Project Kuiper-adjacent Leo Regulus team, requiring 3+ years of professional software development, 2+ years of design/architecture experience, and 2+ years of full-stack experience. The work is explicitly AI-native: designing autonomous agents and orchestration systems that replace manual, cross-domain (tax, accounting, legal, trade, commerce) validation work needed to launch satellite connectivity in 100+ countries.
Step into this interview
Free · a live voice mock calibrated to this exact role
What this interview tests
- Designing multi-agent orchestration systems that run in parallel
- Grounding autonomous agent behavior in domain rules (e.g. accounting, tax, compliance)
- Full-stack system design and architecture judgment
- Reliability and scaling without proportional headcount growth
- Diagnosing and remediating agent-produced errors autonomously
Common question themes
Tell me about a system you built that coordinated multiple independent components or workers in parallel
How would you design a knowledge base that grounds an AI agent's behavior in strict domain rules like accounting or compliance
Walk through a full-stack feature you designed end to end, including the architecture tradeoffs
How do you validate that an autonomous system's output is correct without a human checking every case
Describe how you'd scale a validation process from one country to 100+ without linearly scaling the team
Related interviews
Amazon
Mid
Software Development Engineer II, Amazon Smart Vehicles
Amazon
Mid
Software Development Engineer, Ads AI Core Infra
Amazon
Mid
Data Engineer II, Veritas, Security Tools Foundation
Cohere
Mid
Software Engineer, Agents & Automations
Replit
Senior
Senior Product Engineer, Product Foundry
Affirm
Mid