AI Agent 后台开发工程师 AI Agent Engineer

Igloo

Igloo

Software Engineering, Data Science
Chengdu, Sichuan, China
Posted on May 14, 2025

1. Develop large model applications based on mainstream development frameworks (such as LangChain, LlamaIndex) to create AI solutions for intelligent question answering, knowledge management, and automated processes;

2. Responsible for the design and development of the Agent system, including the implementation of core modules such as task planning, multi-turn dialogue management, and intent recognition;

3. Develop workflows on platforms like Coze/Dify, integrating large model capabilities to achieve business process automation, designing low-latency inference services and asynchronous processing mechanisms;

4. Continuously explore the use of AI capabilities to optimise product experience in various scenarios, proactively innovating practices;

5. Stay updated on relevant cutting-edge technologies, introducing new technologies and solutions to continuously enhance the expressiveness of products and industry competitiveness.

1. Bachelor's degree or above in a related field of Computer Science, with over 3 years of service development experience;

2. Solid coding and development skills, proficient in at least one programming language such as Python or Go;

3. Experience in building complex task scheduling systems, Chatbots, and other related practical experience;

4. Familiar with the entire Agent development process, with practical experience on platforms like Dify/Coze or frameworks such as LangChain/LangGraph;

5. Experience with vector databases, Knowledge systems, and other RAG-related technologies;

6. Good project management and teamwork skills, with excellent analytical problem-solving abilities and a strong sense of responsibility.

The following conditions are considered advantageous:

1. Preference for those with experience in large model algorithms/products;

2. Preference for those with solid back-end development experience;

3. Preference for those with theoretical and practical knowledge in reinforcement learning.