大语言模型Code Agent算法工程师 - Seed Model
Core
Developing an intelligent agent system for code tasks that understands complex code, uses development tools, performs multi-step reasoning, and completes real-world software engineering tasks.
Role type
Senior IC large language model algorithm engineer (Code Agent)
Builds
Intelligent agent systems for software engineering tasks
Domain
Artificial Intelligence / Large Language Models / Software Engineering
Deliverable
production ML models
Required skills
Deep learning frameworks (PyTorch/TensorFlow/JAX), data structures and algorithms, large language model algorithms, reinforcement learning algorithms, data construction, supervised fine-tuning (SFT), reward design, benchmarking, data loop construction
Preferred skills
Research publications in LLM/RL/Agent domains, leading influential projects in LLM/Agent fields
Technologies
PyTorch, TensorFlow, JAX
Responsibilities
Develop algorithms for code understanding, instruction following, and long-horizon task execution; Optimize pre-training and post-training (SFT, RL, Reward); Build evaluation systems and data feedback loops for software engineering scenarios
Seniority
Senior, hands-on IC
