← Back to list
AI Agent Development Engineer
Dongxu Technology
Responsibilities
- Design and develop the core modules of multi-agent system architecture, including a Supervisor-mode multi-Agent scheduling system to optimize task allocation and collaboration logic.
- Build stable, high-availability multi-Agent systems using frameworks such as Hermes and LangGraph, enabling complex workflow orchestration and task coordination.
- Proficiently construct Skills and MCPs; design and implement Agent memory systems and Reflector self-evolution mechanisms to enhance output quality and stability.
- Encapsulate algorithmic modules into production-grade RESTful / gRPC interfaces for backend system integration.
- Establish an LLM-as-Judge evaluation framework and implement content quality gates (Quality Gate).
- Develop GMV attribution logic to deeply align algorithmic performance with business outcomes.
- Optimize multimodal vector retrieval systems to ensure precise recall from brand knowledge bases and industry know-how.
- Enhance video generation quality to improve marketing conversion rates.
- Stay abreast of cutting-edge AI technologies and continuously iterate algorithmic solutions.
- Master prompt engineering, context engineering, and advanced techniques for leveraging domain knowledge.
Requirements
- Deep practitioner in Agentic AI with a strong understanding of Multi-Agent architecture design.
- Proficient in Python, skilled in Skill construction and system development, and experienced with Agent frameworks such as LangGraph / LangChain.
- Familiar with multi-Agent collaboration paradigms like Supervisor mode and Team Agent, with hands-on experience implementing Reflector / Reflection mechanisms.
- Proven experience building stable multi-Agent systems based on Hermes, capable of encapsulating algorithmic modules into robust interfaces for backend consumption.
- Strong grasp of Agent workflows and frameworks, with the ability to rapidly deploy business processes.
- Prior experience in video generation or marketing recommendation projects is preferred.
- Strong coding skills with the ability to quickly translate solutions into production-ready services.
- Preferred Qualifications:
- In-depth understanding of GMV attribution logic, with the ability to correlate technical metrics with business KPIs.
- Familiarity with marketing growth data flywheel mechanisms.
- Systematic knowledge of Agent evaluation methodologies.
- Experience building quality gate systems and multi-level content moderation.
- Practical experience deploying Marketing Agents, particularly in content generation and ad optimization scenarios.
- Production-level deployment capabilities, including experience with high-concurrency, high-availability AI systems.
- Cross-domain confidence assessment skills and expertise in domain adaptation technologies.
About the company
Dongxu Technology is an AI-driven commercial video generation and omnichannel marketing operating system company. Its flagship product, Chorify AI (Tiance Wanying), is positioned as an 'Enterprise-Level AI Digital CGO (Chief Growth Officer), ' replacing traditional marketing planning and video delivery teams through conversational interfaces.
- Industry
- AI Applications & LLMs
- Founded
- 2024-01-29