Skild AI is building the world's first general purpose robotic intelligence that adapts to unseen scenarios. They are seeking a Software Engineer to develop and optimize software infrastructure and tools for training AI models, focusing on building scalable training pipelines and collaborating with researchers.
Responsibilities:
- Develop and maintain robust, scalable, and distributed training pipelines (data preprocessing, training orchestration, and model evaluation) and frameworks for large-scale AI models
- Optimize training processes for performance and resource utilization, ensuring scalability and reliability
- Collaborate with researchers and machine learning engineers to integrate state-of-the-art algorithms and techniques into training pipelines
- Monitor and analyze training, identifying bottlenecks and proposing solutions to improve efficiency and performance
- Ensure the robustness and reliability of the training infrastructure, including automated testing and continuous integration