The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Re-Architecting Data Science at iRobot with Angela Bassa - TWIML Talk #294
Aug 26, 2019
Angela Bassa, Director of Data Science at iRobot, shares her journey from academia to robotics, revealing how she built a data-driven culture at iRobot. She discusses the company's re-architecture and the innovative ways they taxonomize data science. Angela highlights the transition to Python, enhancing team collaboration and product development. The conversation also covers the crucial role of DevOps in integrating hardware, software, and cloud solutions, and the importance of rigorous testing for new technologies like SLAM in robots.
AI Snips
Chapters
Transcript
Episode notes
Demonstrate Quick Wins
- Focus on demonstrating a quick return on investment for data science projects.
- This builds momentum and secures stakeholder buy-in for larger, long-term initiatives.
Iterative Approach
- Starting small with data science allows for early wins and organizational learning.
- This sets the stage for more ambitious, foundational changes later.
iRobot's Re-architecture
- iRobot transitioned from independent product platforms to a modular, shared codebase.
- This change promotes code reuse, faster development, and interoperability across products.