обновлено 5 дней назад
Staff Back-end Engineer (Business Support and Innovation) (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Staff Back-end Engineer (Business Support and Innovation) (AI): LLM Multi-Agent , , with an accent on Agent Function Calling Focus on Agent/RAG Memory Long Context Planning Tool Use
Location: Shanghai, China
Date posted: 30.08.2026
Company
is a global public commerce company building technology and services that simplify shopping, eating, and everyday life.
What you will do
- Lead the architecture and core development of LLM- and multimodal-model-based multi-agent systems.
- Build long-running hosted agents, sandbox execution environments, layered context management, persistent storage, and Function Calling orchestration.
- Deliver complex Agent and RAG systems from solution design through production deployment.
- Develop AI platforms for supply-chain automation, ordering assistance, inventory planning, and enterprise decision support.
- Design and operate highly concurrent, highly available backend services.
Requirements
- End-to-end production experience with complex Agent or RAG systems.
- Knowledge of Memory, Long Context, Planning, and Tool Use, with the ability to turn research into production engineering.
- Experience with one or more AI development frameworks, including LangChain, LangGraph, Deerflow, Claude Code, or Hermes.
- Strong Java or Python skills and experience designing and launching large-scale distributed systems.
- Understanding of asynchronous processing, concurrency, idempotency, rate limiting, design patterns, code quality, and maintainable architecture.
Culture & Benefits
- Entrepreneurial startup culture combined with the resources of a large public company.
- Opportunities to drive new initiatives and innovations in an always-on, high-tech commerce environment.
- Hands-on work with cloud computing, data analytics, machine learning, and generative AI.
Hiring process
- Application review, phone interview, onsite or virtual onsite interview, and offer.
- Interview scheduling and results are communicated by email.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →