大模型算法工程师（Agent方向）-剪映CapCut
Core
Research and develop video editing Agents using multimodal understanding and frontier AI technologies to revolutionize industry efficiency for the CapCut video editing ecosystem.
Role type
Senior IC large model algorithm engineer (Agent direction)
Builds
Multimodal Agents, video editing tools, and AIGC products
Domain
Video editing + Large Language Models + Multimodal AI
Deliverable
production ML models
Required skills
Multimodal understanding, Multimodal RAG, Visual Chain-of-Thought, Multimodal Agents, Reward modeling, Reinforcement Learning, LLM fine-tuning, Prompt/Context Engineering, Tool Use, GUI interaction
Preferred skills
ACM/ICPC/NOI/IOI/Top Coder/Kaggle awards, Published papers on LLMs/Agents, Multi-Agent system experience, AIGC practical experience
Technologies
RL, SFT, LLM architectures, Multi-Agent frameworks
Responsibilities
Research multimodal understanding and Agent technologies; Build datasets and optimize models for vertical video editing scenarios; Implement Agent capabilities like search, coding, and tool use.
Seniority
Senior, hands-on IC
