AI Large Model Management Trainee

Job Description:

• Participate in the design and development of core AI large model-related business and products under the guidance of project managers.

• Master advanced machine learning, deep learning, and large model algorithms and frameworks.

• Engage in code writing and architectural design practices, and assist in project management tasks.

• Receive systematic training and diverse project experience to deeply understand business logic and actively promote new business exploration and implementation.

Responsibilities:

• R&D and application of large language models (LLM), combined with retrieval augmentation (RAG) technology to develop intelligent applications.

• Learn and apply digital human technology, exploring multimodal interactions and virtual image generation.

• Optimize large models under guidance to improve inference speed and reduce deployment costs.

• Explore acceleration strategies for hardware and frameworks like GPU, TensorRT, ONNX Runtime, and achieve end-to-end performance optimization.

• Focus on parallel training and inference of large models in distributed scenarios to enhance scalability and throughput.

• Research and optimize model compression and deployment processes to achieve efficient inference.

• Participate in the development of core modules such as TTS and ASR, optimizing naturalness of voice synthesis and recognition accuracy.

• Collaborate with product and other technical teams to create various landing scenarios like voice interaction and virtual anchors.

• Assist project managers in planning, progress tracking, and risk management to ensure projects are delivered on time and to high standards.

• Participate in code reviews and write technical documentation to promote knowledge sharing and technical accumulation within the team.

• Maintain close communication with business, testing, operations, and other parties to solve various issues that arise in projects.

• Actively participate in the company's management trainee training program, systematically learning business knowledge, project management, professional quality, and other courses.

• Continuously improve programming and architectural design capabilities under guidance, mastering and practicing microservices, containerization, CI/CD, and other management and operational technologies.

• Stay updated with industry trends and technological advancements, daring to trial and innovate, and applying learning to practical work.

Requirements:

• Permanent residents of Hong Kong or persons permitted to work legally in Hong Kong.

• Full-time undergraduate students from universities in Hong Kong, holding bachelor's, master's, or doctoral degrees in STEM-related disciplines from local universities, local universities in collaboration with non-local universities, or specially recognized non-local institutions.

• Solid foundation in mathematics, machine learning, and deep learning, understanding common model structures and training methods.

• Familiar with mainstream deep learning frameworks like PyTorch, TensorFlow, JAX, etc.; some algorithm modeling or engineering experience is preferred.

• Understanding of NLP, speech recognition, TTS, and other fields' basic theories, with data processing and feature engineering capabilities; relevant project or competition experience is a plus.

• Experience in developing with crawlers, langchain, llama_index, etc. is preferred.

• Proficient use of Git or other version control tools, with team collaboration awareness.

At Data Origin (Hong Kong) Limited, we are committed to providing a supportive and rewarding work environment. Our comprehensive benefits package includes:

• Flexible Work Hours

• Snacks and Refreshments

• Birthday Gifts

•Paid Annual Leave

• Medical Insurance

• Additional Allowances