Design and implement agent-based systems that operate reliably in production Build orchestration logic for multi-step and multi-agent workflows Define agent boundaries, responsibilities, and failure handling strategies Make architectural trade-offs balancing accuracy, latency, cost, and reliability Build and maintain AI-backed backend services using Java and/or TypeScript Integrate LLMs into existing APIs, workflows, and data pipelines Ensure AI systems respect authentication, authorization, and access controls Implement guardrails to control AI behavior and prevent unsafe or misleading outputs Design evaluation approaches for AI outputs and agent behavior Implement logging, monitoring, and alerting for AI systems Track and clearly communicate AI limitations, failure modes, and operational risks Manage rate limiting, cost controls, and usage visibility Ensure AI usage aligns with security, privacy, and compliance expectations Review and harden AI-assisted code before production release Guide other engineers on safe and effective AI integration Push back when AI is proposed where it introduces more risk than value Work closely with the Engineering Manager on system design and delivery planning Coordinate with backend and frontend engineers to integrate AI capabilities cleanly Contribute to technical discussions, design reviews, and production readiness efforts