R&D Engineer - Expert

• Enhance R&D efficiency through AI technology application and implementation
• AI Application Development and Project Delivery
• Lead AI project requirements analysis, technical solution design, and product delivery
• Complete fine-tuning, alignment, and inference optimization for general models, embedded models, and inference models
• Build efficient and usable Prompt Engineering workflows to improve model task performance
• Multi-Agent Systems and Framework Applications
• Implement multi-agent workflow orchestration using frameworks like LangChain, LangGraph, and MCP
• Design and implement key reasoning paradigms (ReAct, CoT, ToT) to improve agent system responsiveness and controllability
• Understand agent platforms such as Coze, FastGPT, and Dify
• Knowledge Retrieval and RAG Systems
• Build document knowledge retrieval systems based on vector databases (Milvus, FAISS, Chroma, etc.)
• Design RAG architecture solutions to enable context-enhanced interactions with large language models (e.g., ChatGPT, DeepSeek)
• Improve document recall quality and reasoning relevance
• Technical Research and Capability Development

Track AI technology trends (model alignment, multimodality, multi-agent systems, etc.), regularly complete technical research, develop application prototypes, or deliver technical presentations

• Must Have
• Strong self-motivation and continuous learning ability, passionate about AI, attentive to cutting-edge technologies, industry trends, and business challenges, capable of rapid hands-on experimentation and post-mortem analysis
• Master's degree or higher in Computer Science, Artificial Intelligence, Electronics, Information Technology, or related fields, or equivalent engineering experience
• Proficient in Python or at least one backend language (e.g., Go/Java/C++), with containerization (Docker) skills, familiarity with CI/CD pipelines, and foundational MLOps practices
• Expertise in at least one specialized AI domain with practical experience and demonstrable project outcomes:

o Hands-on experience applying Prompt Engineering, LangChain, or RAG frameworks

o Familiarity with Multi-Agent system architecture and hands-on experience in orchestration using LangGraph or MCP

o Proficiency in selecting and integrating vector databases (e.g., Milvus, Chroma, FAISS)

• Strong English reading/writing and online communication skills to collaborate effectively with overseas AI leaders on goal setting, solution discussions, post-mortems, and documentation
• Results-oriented mindset with ability to decompose ambiguous problems into deliverable milestones, emphasizing system stability, observability, and maintainability
• Nice to Have
• Background in ATE or semiconductor industry, understanding of test flows, test programs (TP), Shmoo/Waveform analysis, yield and anomaly management; familiarity with platforms like 93K is a plus
• Experience with edge or on-premises deployments, familiarity with GPU/CPU acceleration, model quantization and distillation, and optimization techniques for resource-constrained environments
• Participation in or leadership of cross-regional R&D collaboration projects, with practical experience in roadmap development, milestone decomposition, and project execution