4 hours ago
London, United KingdomStaff+
Responsibilities
- Lead Gaia’s post-training and closed-loop pipeline, including fine-tuning, alignment, experimentation, and targeted data curation.
- Improve long-horizon autoregressive generation so Gaia produces stable, coherent, and steerable rollouts.
- Contribute to model architecture and training-strategy decisions across pre-training, post-training, and application layers.
- Translate post-training improvements into measurable downstream impact with research, applications, simulation engineering, and cloud/infrastructure teams.
- Provide technical leadership through mentorship, reviews, and setting high engineering and research standards.
- Improve inference optimization, deployment readiness, reliability, and performance under relevant constraints.
Requirements
- Hands-on experience post-training or fine-tuning large-scale language, video, or other foundation models.
- Experience with world models, autoregressive generation, and long-horizon generation.
- Experience with diffusion or flow models and an understanding of 3D vision.
- Strong understanding of model architecture and ability to contribute to architecture and training decisions.
- Strong hands-on engineering skills with modern ML stacks, including PyTorch, debugging, and performance- and reliability-minded development.
- Typically 5+ years of relevant industry experience.
- Advanced degrees are valued, but substantial applied experience is important.
- Experience with inference optimization, large-model deployment under latency or compute constraints, data or training pipelines, distributed training, efficiency, reliability, and technical leadership is desirable.
Benefits
- Full-time position based in London, UK, with a hybrid working arrangement combining office/workshop and home working.
- Core working hours with flexibility to determine a schedule that works for the individual and team.
- Inclusive interview experience with accommodations or adjustments available on request.
Tech Stack
Categories
AI ResearchML Engineering
About Wayve
Wayve builds end-to-end autonomous driving software—the vehicle-agnostic Wayve AI Driver—that runs on onboard compute and native sensors, licensed to automakers and fleet operators. Its platform spans ADAS and higher autonomy (L2+/L3 to robotaxi) and is designed to generalize across vehicle types and geographies. Founded in 2017 and headquartered in London, it tests its models across Europe, North America, and Japan, with a U.S. base in Sunnyvale, CA.