Generative AI Agent Engineer - Agent Design & Performance

About Chryselys

Role Summary

Responsibilities

• Design agent architectures, workflows, instructions, decision logic, and collaboration patterns.
• Define and build high-quality tools, including tool schemas, inputs, outputs, permissions, error handling, and user feedback.
• Develop agents that can plan, reason, retrieve information, call APIs, and complete multi-step tasks reliably.
• Optimize agent performance across accuracy, task completion, latency, cost, consistency, and user experience.
• Design memory, context, retrieval, and summarization strategies that improve agent effectiveness.
• Create evaluation frameworks, test scenarios, benchmarks, and quality metrics for agent behavior.
• Analyze agent traces and failures to identify issues with prompts, tools, routing, context, or orchestration.
• Implement guardrails for safety, security, hallucination reduction, authorization, and human-in-the-loop escalation.
• Build multi-agent workflows and agent-to-agent interactions where appropriate.
• Collaborate with application and platform engineers to integrate agents into production systems.
• Establish reusable patterns and best practices for agent development and continuous improvement.

Qualifications

• 3+ years of experience in software engineering, machine learning, or GenAI application development.
• Hands-on experience designing and deploying LLM-powered agents or agentic workflows.
• Strong understanding of prompting, tool calling, structured outputs, planning, RAG, memory, and context management.
• Experience designing APIs and tools for agents, with attention to usability, reliability, permissions, and clear contracts.
• Proficiency in Python, TypeScript, or a comparable programming language.
• Experience with agent frameworks such as LangGraph, LangChain, or similar technologies.
• Familiarity with AWS and preferably AWS AgentCore.
• Experience with agent evaluation, observability, tracing, experimentation, and performance optimization.
• Strong understanding of distributed systems, APIs, asynchronous workflows, and production software practices.

Preferred Qualifications

• Experience with multi-agent systems and A2A or MCP protocols.
• Experience building agents for complex enterprise or regulated-industry workflows.
• Familiarity with model routing, prompt versioning, fine-tuning, guardrails, and human-in-the-loop patterns.
• Strong understanding of identity, authorization, data privacy, and secure tool execution.
• Experience translating business processes into reliable agent workflows and measurable outcomes.