AI Researcher – Audio & Multimodal AI

Research centre

Role overview

Research leadership

• Contribute to and progressively lead multimodal foundation-model programs, especially speech/audio LLMs and generative audio systems.
• Help define research priorities and align them with centre strategy.
• Support and progressively mentor junior researchers and engineers.
• Engage in decisions on research roadmaps, model development priorities, and team initiatives.

Scientific and technical work

• Design, implement, and rigorously evaluate novel architectures and training algorithms for audio and multimodal generative systems.
• Participate in dataset curation, model training, fine-tuning, scaling, and systematic performance evaluation.
• Contribute to large-scale data processing and experiment management across distributed training environments.
• Drive open-source releases and publish findings in leading venues.

Collaboration and responsible AI

• Collaborate with internal centres, industry, academia, and end-user communities.
• Present outcomes in publications, reports, demonstrations, and conference presentations.
• Contribute to proposals for research programs and strategic initiatives.
• Follow ethical standards, responsible AI practices, and AI safety guidelines; help establish shared infrastructure and reusable components.

Required qualifications and skills

• PhD in Computer Science, Artificial Intelligence, Electrical Engineering, or related field is strongly preferred; a Master's degree with demonstrated research experience and output is an alternative.
• Expertise or strong research interest in audio AI, including one or more listed speech, audio, music, or multimodal areas.
• Research output through publications, preprints, thesis work, or open-source contributions demonstrating rigor and originality.
• Understanding of modern AI architectures, training paradigms, and evaluation methodologies.
• Hands-on experience in at least one of audio dataset curation, model training, or audio generation evaluation.
• Proficiency in Python and experience with PyTorch or TensorFlow.
• Good written and verbal communication and ability to present research clearly.

Additional strengths

• AI, audio, speech, music, or multimodal learning patents are advantageous.
• Relevant publications in speech, audio, music, or general AI/ML venues and journals are advantageous.
• Experience in ML or audio competitions such as Kaggle or DCASE Challenge is advantageous.