Sr. Software Dev Engineer, Leo AI Foundations

About the role

This role is for a Sr. Software Development Engineer who will design, implement, and operate globally distributed systems that enable Amazon Leo to achieve low single-digit-second query responses within a near real-time analytics layer or lakehouse, with a primary focus on agentic AI capabilities for autonomous operational intelligence and system optimization.

Responsibilities

• Architect multi-agent systems that continuously analyze telemetry data from satellites, ground gateways, and customer terminals to detect anomalies, predict failures, and autonomously recommend corrective actions
• Develop intelligent agents capable of reasoning across complex distributed systems to identify root causes of operational issues with minimal human intervention
• Build agentic workflows that autonomously triage system anomalies, escalate critical issues based on severity and impact, and generate actionable insights for operations teams
• Design agent-based frameworks that coordinate workflows across the constellation, enabling collaborative problem-solving between autonomous agents, ground systems, and operations teams
• Implement reinforcement learning-based agents that optimize system performance parameters in real-time based on environmental conditions, network demand, and operational constraints
• Develop natural language interfaces allowing operations teams to query system health status, request analyses, and receive AI-generated recommendations through conversational interactions
• Build RAG systems that combine real-time telemetry with historical operational data, technical documentation, and knowledge bases to provide context-aware insights
• Design hybrid search strategies combining dense vectors with sparse representations for optimal semantic retrieval across operational patterns and system behaviors
• Design and implement evaluation frameworks to measure agent performance, accuracy, and reliability across diverse operational scenarios
• Build automated testing pipelines for agent behavior validation, including unit tests, integration tests, and end-to-end scenario testing
• Establish metrics and monitoring systems to track agent decision quality, response times, and operational impact
• Create feedback loops that continuously improve agent performance through reinforcement learning and human-in-the-loop validation
• Arcitect and implement a scalable, cost-optimized S3-based Data Lakehouse that unifies structured and unstructured data from disparate sources across the Leo constellation
• Establish metadata management with automated data classification and lineage tracking to support both analytical queries and AI retrieval patterns
• Design and enforce standardized data ingestion patterns with built-in quality controls and validation gates for satellite telemetry, ground station metrics, and customer terminal data
• Arcitect and implement a scalable, cost-performance-optimized OLAP-based analytics layer capable of achieving low single-digit-second query responses for near real-time analytics
• Lead the design of semantic data models that balance analytical performance with AI retrieval requirements
• Implement cross-domain federated query capabilities with sophisticated query optimization techniques
• Arcitect a centralized metrics repository that becomes the source of truth for all Leo operational metrics
• Design extensible metrics schemas that support complex analytical queries while optimizing for AI retrieval patterns
• Implement robust data quality frameworks with staging-first policies and automated validation pipelines
• Develop intelligent orchestration for metrics generation workflows with comprehensive audit trails
• Arcitect a globally distributed vector database infrastructure capable of managing billions of embeddings with consistent sub-100ms retrieval times
• Design and implement hybrid search strategies for optimal semantic retrieval across operational documentation, telemetry patterns, and system knowledge bases
• Arcitect automated compliance validation frameworks ensuring data handling meets Amazon's security standards and export control requirements

Qualifications

• 5+ years of non-internship professional software development experience
• 5+ years of programming with at least one software programming language experience
• 5+ years of leading design or architecture (design patterns, reliability and scaling) of new and existing systems experience
• Experience as a mentor, tech lead or leading an engineering team