Job description
Responsibilities
1. Image and Video Generation: Build workflows for image and video generation based on ComfyUI, completing tasks such as text-to-image, text-to-video, image-to-video, and video frame interpolation, supporting batch content production.
2. Character Consistency and Scene Optimization: Implement head swapping, face swapping, and character consistency control in videos, optimizing face alignment, feature fusion, and temporal coherence, fixing issues like character deformation and scene flickering.
3. Workflow Documentation and Automation: Document prompt templates, LoRA configurations, and workflow SOPs, using Python for debugging, batch processing, and business interface integration to enhance generation efficiency and control computing costs.
4. Business Effect Iteration: Continuously optimize content quality based on business feedback, collaborating to generate and deliver AI character-related image and video content.
Requirements
1. 1-3 years of practical experience in AIGC image or video generation, capable of independently building workflows, optimizing effects, and delivering content, with no educational restrictions.
2. Proficient in using ComfyUI, able to independently build and debug complex workflows; familiar with models such as Stable Diffusion, FLUX, LTX series, and proficient in the application of ControlNet and LoRA.
3. Familiar with face detection, alignment, and image fusion methods, understanding tools like MTCNN and RetinaFace, capable of addressing character consistency and video coherence issues.
4. Basic Python programming skills, able to perform workflow debugging, interface calls, and batch task processing.
5. Aesthetic judgment and content quality assessment skills, able to use tools like Jianying or Premiere for basic editing and post-processing.
6. Must provide works or case studies participated in, detailing the models used, workflows, personal responsibilities, and specific problems solved.
Bonus Points
- Experience in video face swapping, digital humans, or batch production of AI character content.
- Experience in LoRA training, dataset organization, or GPU deployment and performance optimization of open-source models.
- Experience with LLM applications, able to integrate Prompt, Agent tool calls, or RAG with content generation workflows.
- Experience in developing custom nodes for ComfyUI, TTS voice generation, or audio-visual synchronization.
- Familiarity with anime, web literature, AI companionship, or interactive entertainment content, able to understand related business needs.