Описание
The Enterprise Agent Development Platform is a production-grade, cloud-native ecosystem that enables engineering teams to define, orchestrate, deploy, and observe AI agents at scale. It standardizes agent development across the organization using LangGraph and Strands Agents on AWS AgentCore Runtime, spanning agent framework design, runtime architecture, marketplace integration, CI/CD automation, and enterprise-grade observability. The platform aims to reduce agent development from months to days while enforcing consistent security, quality, and governance standards.
Задачи
•Integrate AgentCore Memory into platform agent templates and helper libraries
•Implement write-on-turn functionality to store session events
•Implement retrieve-on-turn functionality to fetch relevant context
•Design and build memory namespace architecture for per-agent isolation and per-user scoping
•Develop Python helper libraries to support memory-enabled agents across the platform
•Enable memory activation through configuration rather than custom code in agent templates
•Use AWS AgentCore Memory API to create, query, update, and delete memory records
•Build and maintain API integrations for stateful services
•Manage session persistence and context continuity across agent interactions
Требования
•7+ лет опыта backend-разработки с AI/ML системами
•Практический опыт интеграции agent memory в продакшн agentic AI проекте, например AWS AgentCore Memory API (write-on-turn/retrieve-on-turn), проектирование memory namespace или конфигурация памяти агентных шаблонов
•Экспертиза в разработке Python SDK
•Навыки интеграции API для stateful-сервисов
•Владение управлением сессиями и context persistence
•Английский на уровне B2 и выше
Будет плюсом
•Знакомство с AWS AgentCore Memory SDK
•Знание интеграции памяти LangChain
•Опыт разработки агентных фреймворков (LangGraph, Crew)
•Понимание обработки контекста в разговорном AI
Условия
Условий в вакансии нет
