大模型算法工程师(27届校招)_BCSC
Core
Build core algorithms for an AI Agent service platform, focusing on runtime frameworks, context engineering, multi-agent collaboration, and long-term planning.
Role type
Large Model Algorithm Engineer (Campus Recruitment)
Builds
AI Agent service platform, Agent runtime frameworks, evaluation systems, and multi-agent collaboration solutions.
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
Transformer architecture, pre-training and fine-tuning paradigms, PyTorch, Hugging Face Transformers, alignment algorithms (SFT, DPO, PPO), Agent concepts (ReAct, Tool Calling), model inference engines (vLLM), research paper reproduction
Preferred skills
Model service deployment, top-tier conference publications, open-source contributions
Responsibilities
Iterate and optimize Agent runtime frameworks and context engineering; build and maintain Agent performance evaluation frameworks; design and enhance multi-agent collaboration systems for automation scenarios; research and implement long-term Agent planning and business skills; execute large model post-training optimization (SFT, DPO, ORPO) in business scenarios; establish end-to-end evaluation systems from experiment to launch.
Seniority
Junior, Campus Recruitment (Master's degree)
