广告算法工程师（出价和竞价方向）-中国交易与广告
Core
Build an end-to-end generative bidding and auction system that uses sequence modeling to generate optimal bidding actions directly, maximizing advertiser value under budget constraints while replacing traditional win-rate estimation with AI-driven real-time decision-making.
Role type
Senior IC machine-learning engineer (generative bidding and auction)
Builds
End-to-end generative bidding and auction systems for ad trading
Domain
Online advertising, machine learning, reinforcement learning
Deliverable
production ML models
Required skills
Python, Java, deep learning frameworks (PyTorch, TensorFlow), reinforcement learning (DQN, A3C, PPO, CQL), Transformer architectures, large language models, sequence modeling, reward function design, sample selection, exploration mechanisms
Preferred skills
Search, advertising, and recommendation scenarios experience, Agentic AdsBrain, Decision Transformer (DT), GRPO, multi-agent collaboration
Technologies
MPC, Decision Transformer, GRPO, Agentic AdsBrain, DQN, A3C, PPO, CQL, PyTorch, TensorFlow
