AI算法工程师（Agent Harness研究）-即梦
Core
Design and build algorithmic capabilities for the Agent Harness in Dreamina's image, video, and multimodal creation scenarios, enabling proactive agents to plan, execute, and optimize complex creative tasks.
Role type
Senior IC machine-learning engineer (agentic systems)
Builds
Agent Memory, Agent Skills, multimodal context organization, and proactive agent modules for creative workflows
Domain
AIGC, multimodal generation, agentic AI
Deliverable
production ML models
Required skills
LLM/VLM, Tool Use, Function Call, task planning, workflow orchestration, Python, algorithmic experimentation, data analysis, effect evaluation
Preferred skills
Human-in-the-loop design, agent trajectory evaluation, automated diagnosis, failure case attribution
Technologies
Python, LLM, VLM, AIGC
Responsibilities
Explore creative planning, task decomposition, execution strategy selection, and multi-turn function calls; Design human-in-the-loop collaboration mechanisms; Collaborate with engineering, model, evaluation, and product teams to deploy algorithms.
