- Train, fine-tune, and ship computer-vision models for thermal anomaly detection, defect detection, and object detection.
- Develop and maintain the MLOps backbone, including experiment tracking, dataset versioning, model registry, and deployment pipelines.
- Execute the full experimental loop from dataset curation and training to error analysis.
- Architect solutions for large-scale spatial context models that exceed standard fixed-resolution capabilities.
- Integrate models into the product end-to-end to ensure real-world performance.
- Prioritize AI development approaches based on business impact and product requirements.
PythonDeep LearningMLOps+1 more