算法工程师（音乐智能理解方向） - TikTok研发
Core
Develop algorithms for music audio understanding and generative AI to enable next-generation intelligent music experiences on TikTok.
Role type
Senior IC machine-learning engineer (audio/music)
Builds
TikTok music features and audio intelligence capabilities
Domain
Music technology + Generative AI
Deliverable
production ML models
Required skills
Python, PyTorch, Transformer architectures, Diffusion models, audio signal processing, deep learning
Preferred skills
LLMs, CLAP, Encodec, DiffSinger, audio/music generation models, top-tier conference publications (ISMIR, ICASSP, Interspeech, NIPS), music theory knowledge, composition skills, instrument proficiency
Technologies
PyTorch, Transformer, Diffusion, Neural Audio Codec, TTS, SVS, Audio-LLM
Responsibilities
Develop algorithms for music feature extraction, genre/emotion recognition, automatic tagging, lyrics-audio alignment, and structure segmentation; Research audio neural coding, singing/voice synthesis, audio super-resolution, and quality enhancement; Build audio datasets, perform feature preprocessing, generate synthetic data, and establish evaluation metrics.
Seniority
Senior, hands-on IC