Technical Staff
Kyutai · Paris, FR
Research instituteonsite
At a glance
Highlights
- Build impact-driven projects in generative modeling, video and world models.
- Publish reproducible research with open models, code and public demos.
- Join team behind Moshi, Helium-1, Hibiki and other groundbreaking models.
What they offer
- Highly collaborative, positive and stimulating environment with elite team
- Commitment to open science with publications, open-source models and demos
- Research freedom and ample compute resources for ambitious directions
- Competitive compensation and benefits package
- Assistance with relocation and immigration procedures if needed
Ideal candidate
Curiosity-driven researcher or engineer with deep ML experience who thrives on autonomous, high-impact open science and wants to push frontier multimodal AI with an elite team.
Requirements
- Hands-on experience with multi-GPU model training and large-scale clusters
- Experience in speech modelling, efficient LLMs, video understanding, world models or related areas
- PhD in ML-adjacent field or 4+ years of experience in ML engineering
- Track record of key contributions to impactful projects
- Extreme focus, dedication and taste for autonomous curiosity-driven research
- visa sponsorship
- Assistance with relocation and immigration procedures if needed
Lab profile
About Kyutai
Non-profit open research lab building multimodal, compact frontier AI models.
What they build
- Moshi full-duplex voice AI
Known for
- Open frontier AI research
- Multimodal, adaptable compact models
- Real-time speech and language modeling
- Video understanding and world models
Key facts
- Founded in 2023
- Based in Paris
- Non-profit lab
Posted · Last seen