豆包大模型算法工程师（AIGC Agent方向） - 火山方舟
Core
Optimizing AIGC Agent chains and enhancing generation models (Seedream, Seedance) for improved task completion and user experience.
Role type
Senior IC Machine Learning Engineer (AIGC Agent & Model Post-training)
Builds
AIGC Agent systems, generation models, and AI application services for enterprise clients.
Domain
Artificial Intelligence / Large Language Models / AIGC
Deliverable
production ML models
Required skills
LLM/VLM training and post-training, Reinforcement Learning (RFT, RLHF, GRPO), Online Policy Distillation, SFT data construction, Agent planning and reasoning, Context management, Multi-Agent collaboration, Self-evolving Agent capabilities
Preferred skills
Image/video generation model optimization (Diffusion, VAE, Transformer-based), Publications in top AI conferences (NeurIPS, ICML, etc.), Competitive programming or AI contest achievements
Technologies
RFT, RLHF, GRPO, OPD, SFT, Diffusion, VAE, Transformer-based Generation, Multi-Agent systems
Responsibilities
Optimize generation models in AIGC Agent chains, conduct model post-training and capability enhancement, build AIGC Agent system capabilities, analyze complex creation scenarios to close optimization loops, deploy frontier model capabilities into real business scenarios
Seniority
Senior, hands-on IC