1 day ago
Sunnyvale, CA, USAStaff+
Base Salary
$370k - $407k/yr
Responsibilities
- Build, train, and fine-tune scene-understanding models for offline measurement, adapting on-vehicle architectures and Wayve Foundation Models.
- Improve model accuracy and generalization across vehicle platforms, geographies, and driving conditions by diagnosing failure modes.
- Use larger compute budgets, larger models, bidirectional temporal context, and multi-task or joint representation learning in offline modeling.
- Benchmark models, define quality bars, analyze errors, and use metrics to guide iterative development.
- Define ground truth and correctness criteria across a complex driving taxonomy and create automated benchmarks and statistically defensible evidence.
- Ensure results can support validation pipelines and safety cases at scale across the product portfolio.
- Collaborate with on-vehicle modeling, evaluation, data curation, and simulation teams, while setting technical direction and mentoring others.
Requirements
- At least 5 years of ML engineering experience, including training and shipping deep learning models in production.
- Hands-on experience training modern computer vision models, including transformer-based and multimodal or VLM architectures, for detection, segmentation, classification, or scene understanding.
- Experience adapting or fine-tuning pretrained or foundation models and training shared representations across multiple tasks or objectives.
- Proficiency in Python and ML frameworks, especially PyTorch, with strong software engineering practices and experience with large-scale training.
- Staff-level technical leadership, including setting direction, leading cross-functional work, and mentoring without formal line management.
- Ability to define and interpret metrics that demonstrate whether models are improving.
- Experience with 3D scene understanding, geometric and semantic perception, or large-scale semantic enrichment of driving scenes is desirable.
- Experience with offline modeling, auto-labeling, model distillation, temporal or world models, or other offboard modeling approaches is desirable.
- Prior autonomous vehicle or robotics experience with deployment and closed-loop validation on physical systems is desirable.
- Experience with fleet-scale data and large-scale distributed training infrastructure is desirable.
Benefits
- Full-time position based in the Sunnyvale office.
- Hybrid working policy combining office and workshop collaboration with working from home.
- Competitive equity package.
- Inclusive interview experience with accommodations available upon request.
Categories
