AI应用技术Leader-集团信息系统
Core
Synthesize, clean, and curate pre-training data for large language models; optimize reward models and reinforcement learning algorithms to improve model reasoning and alignment.
Role type
Senior IC machine-learning engineer (RL & LLM alignment)
Builds
Production LLMs with enhanced reasoning capabilities and reward models
Domain
Artificial Intelligence / Large Language Models / Reinforcement Learning
Deliverable
production ML models
Required skills
Data synthesis, data cleaning, reinforcement learning, reward modeling, SFT, algorithm design, Python programming, model architecture understanding
Preferred skills
Reinforcement learning project experience, data-driven work in recommendation systems, rapid prototyping, competitive programming (NOI/ACM)
Responsibilities
Develop data synthesis pipelines to improve pre-training data quality; Optimize reward models and RL algorithms for high-value data feedback loops; Collaborate on SFT stages to address discrimination capability gaps; Organize annotation efforts for reward model training.
