Deep Learning Engineer (음성 인식 및 wake word detection 개발)
Core
Designing and optimizing speech AI models for accurate voice understanding and wake word detection in diverse vehicle environments to power LLM-based AI Agents.
Role type
Senior IC deep learning engineer (speech recognition & wake word detection)
Builds
STT and Wake Word Detection models, on-device and server-side speech applications for Linux and Android
Domain
Automotive mobility, speech AI, on-device machine learning
Deliverable
production ML models | product features
Required skills
STT model design, Wake Word Detection (WWD) model design, on-device application development (Linux/Android), hardware-efficient model optimization, C/C++/Python programming, deep learning frameworks (PyTorch/TensorFlow/Keras/Caffe)
Preferred skills
SSL (Self Supervised Learning) research, AED (Attention based Encoder-Decoder) research, multi-lingual STT research, language model research for speech recognition, open-source contribution, unit testing and CI/CD practices
Technologies
PyTorch, TensorFlow, Keras, Caffe, Linux, Android, C, C++, Python, Git
Responsibilities
Design STT acoustic and language models and build speech databases; Design WWD models and build speech databases; Develop batch and streaming speech applications for server and on-device; Develop and optimize hardware-efficient models
Seniority
Mid-level to Senior, hands-on IC