Character.AI06.03.2026
Research Engineer, Multimodal
Полная занятостьУдалёнка
Обязанности
- 01Lead fine-tuning and continued training of video generation models, including image-to-video and joint audio-visual generation
- 02Design and experiment with novel model architectures for multimodal generation, including multimodal conditioning (voice, structured text, reference images)
- 03Leverage techniques such as LoRA, RLHF, and full-parameter fine-tuning to improve model quality across diverse visual scenarios
- 04Design and build large-scale data pipelines and automated annotation workflows to support continuous model improvement
- 05Explore model compression, inference acceleration, and serving optimizations to enable efficient real-time video processing at scale
Требования
- 01Strong passion for pushing the boundaries of visual AI, with a self-driven, hands-on approach to solving complex technical problems
- 02Proficient in PyTorch with end-to-end experience across data processing, model training, and deployment
- 03Solid understanding of video and image generation architectures, including diffusion models, DiT, ControlNet, and SOTA video generation models
- 04Experience with multimodal model training, including working with audio, vision, and language modalities together
- 05Experience with distributed training tools (FSDP, DeepSpeed, etc.)
- 06Experience with large-scale data processing, dataset construction, and automated data cleaning
Условия
- 01Join a unicorn startup with rapid growth and innovation in AI
- 02Work on cutting-edge multimodal AI models shaping user experiences for millions
- 03Collaborate with a diverse and inclusive team in a dynamic environment
- 04Opportunities to publish research at top-tier conferences (NeurIPS, ICLR, CVPR, etc.)