Key Responsibilities:

Develop core Agent AI capabilities, including task planning, task decomposition, tool use, memory systems, state management, and result verification.

Design and implement Agentic Workflows, such as scheduled tasks, automated executions, periodic report generation, auto‑retrieval, and auto‑summarization.

Contribute to multi‑agent system development, enabling collaboration among Planner, Executor, Reviewer, Memory, and other specialized agents.

Support “multi‑avatar” (multiple digital persona) applications, where several agent instances run in parallel, each with its own context and task objectives.

Integrate LLM APIs, local models, RAG knowledge bases, and business systems to improve task completion rates and system stability.

Build observability and control mechanisms including agent logs, task tracing, failure retries, human‑in‑the‑loop review, and security guardrails.

Participate in agent evaluation and monitoring – track task success rate, hallucination rate, latency, cost, and user satisfaction.

Stay up‑to‑date with leading Agent/LLM frameworks (LangChain, LangGraph, LlamaIndex, AutoGen, CrewAI, etc.) and make informed technology choices based on business needs.

Requirements:

1–3 years of hands‑on experience in AI engineering, backend development, NLP, machine learning, or LLM application development.

Proficient in Python with strong backend engineering skills.

Familiar with at least one Agent or LLM application framework, e.g., LangChain, LangGraph, LlamaIndex, AutoGen, or CrewAI.

Solid understanding of Prompt Engineering, Function/Tool Calling, RAG, Embeddings, and vector databases.

Experience with scheduled jobs, async task queues, automation workflows, or SaaS systems is a plus.

Experience with Chinese language corpora, Chinese‑language agents, or business scenarios targeting the China market is a plus.