AI Engineer

Responsibilities and Skill Requirements

1. Project experience and core delivery capability: 2-3 years of full-time experience in LLM / AI applications, with hands-on, end-to-end experience delivering enterprise-grade RAG and intelligent Agent projects, from development through to online optimization. Able to independently own the development of core modules and drive project iterations to delivery.
2. Agent engineering framework capability: Proficient in the core principles of intelligent Agents, with experience implementing and iteratively optimizing an Agent Harness engineering framework. Skilled in developing core modules such as task orchestration, tool calling, multi-turn memory, autonomous decision-making, and performance evaluation. Participates in building standardized, reusable Agent development and deployment workflows to support complex intelligent business scenarios.
3. Enterprise RAG engineering and optimization: Deep expertise in full-pipeline RAG engineering for complex scenarios. For use cases involving massive knowledge bases, long documents, cross-domain Q&A, and high concurrency, able to implement optimizations in chunking, retrieval, re-ranking, and prompt engineering. Able to independently resolve core production issues such as model hallucination, inaccurate retrieval, and response latency.
4. Data architecture and database capability: Proficient with mainstream vector databases such as PostgreSQL, Milvus, and FAISS, and skilled in architecture design, index tuning, and data governance for conventional business databases. Able to build stable, efficient data pipelines for LLM applications and complete the data feedback loop.
5. Full-stack AI engineering capability: Mature AI engineering practices, including proficient use of Git for code management and team collaboration, and Docker for containerized deployment and operations. Familiar with building and maintaining CI/CD pipelines to automate build, testing, and continuous delivery, ensuring scalable and standardized project iteration.
6. Efficient development with Vibe Coding: Proficient in the Vibe Coding paradigm, skilled at using AI coding tools for code generation, feature development, debugging, and refactoring, while balancing rapid iteration with clean code quality and maintainability.
7. Problem-solving and iteration capability: Solid Python programming foundation and proficiency with mainstream LLM application frameworks such as LangChain and LangGraph. Familiar with LLM plugins, toolchains, and application orchestration patterns. Able to independently break down business requirements, produce technical solutions, and troubleshoot production issues, continuously improving the performance and business value of LLM applications.
8. Bilingual communication and cross-border collaboration: Fluent in both English and Chinese (spoken and written), able to work closely with team members in China, including technical discussions, documentation, and project coordination.

Nice to Have

1. Ability to research cutting-edge LLM technologies, keeping up with the latest developments in LLMs, multimodal models, Agents, and RAG, and able to produce research proposals and innovative implementation ideas tied to business needs.
2. Experience delivering complex multi-Agent collaboration systems, enterprise intelligent assistants, or large-scale knowledge-based Q&A systems is preferred.
3. Familiarity with performance optimization techniques such as LLM fine-tuning, quantization, inference acceleration, and lightweight deployment is preferred.
4. Experience applying AI in the biomedical or a related life sciences field is preferred, including familiarity with biomedical data and an understanding of the relevant regulatory requirements.
5. Bachelor's degree or above in Computer Science, Artificial Intelligence, Software Engineering, or a related field is preferred.

Why Join Us

1. A core new LLM project where you will have full ownership of building the Agent and RAG engineering foundation, with plenty of room for growth and deep involvement in core technical development and project iteration.
2. A cutting-edge tech stack focused on the most important areas of AI deployment today, with little repetitive legacy work, and deep focus on Agent engineering and LLM application delivery.
3. A strong technical culture that encourages innovation and exploration, and supports technology research and knowledge building.