Prompt Engineer

As a Prompt Engineer in our Artificial Intelligence/Product Development/Data Science department, you will utilize your creativity, strong logical thinking, and expertise in large language models (LLMs) to assist in the establishment of an efficient and precise prompt system. Your role will involve driving the implementation of generative AI technology in our products, collaborating closely with machine learning engineers, product managers, designers, and researchers to enhance model response quality and elevate user interaction experiences. Your responsibilities will include designing, testing, and optimizing prompt templates for prominent LLMs like GPT-4, Claude, and Gemini. You will analyze model output, identify potential issues, and iterate on optimizations. Developing prompt libraries, establishing version control mechanisms, and integrating prompt designs into specific application scenarios are crucial aspects of the role. Staying updated on cutting-edge prompt engineering technologies and industry best practices is essential, as well as contributing to prompt data generation for fine-tuning and reinforcement learning. Furthermore, you will be instrumental in crafting prompt test evaluation processes, creating technical documents, and sharing prompt design experiences across teams. The ideal candidate should possess familiarity with mainstream LLMs, a solid foundation in natural language processing or linguistics, and practical experience in prompt engineering and few-shot learning. Proficiency in a programming language, exceptional logical reasoning and writing skills, and a good command of English are required. Bonus points will be awarded for experience in building AI assistants, familiarity with relevant tools, optimization expertise, and knowledge of model governance strategies. In return, we offer opportunities to work on cutting-edge generative AI products, a collaborative interdisciplinary environment that encourages exploration and experimentation, a flexible office system, competitive compensation, growth opportunities, and access to extensive real data, robust model resources, and GPU computing capabilities.,