Machine Learning Platforms at Uber with Mike Del Balso - TWiML Talk #115
Analysis
This podcast episode from Practical AI features an interview with Mike Del Balso, Product Manager for Machine Learning Platforms at Uber. The discussion centers on the challenges and best practices for implementing machine learning within organizations. Del Balso highlights common pitfalls such as inadequate infrastructure for maintenance and monitoring, unrealistic expectations, and the lack of appropriate tools for data science and development teams. The interview also touches upon Uber's internal machine learning platform, Michelangelo, and the open-source distributed TensorFlow system, Horovod. The episode concludes with a call to action for listeners to vote in the #MyAI Contest.
Key Takeaways
- •Organizations should prioritize proper infrastructure for machine learning maintenance and monitoring.
- •Realistic expectations are crucial for successful machine learning implementation.
- •Providing data science and development teams with the right tools is essential.
“Mike shares some great advice for organizations looking to get value out of machine learning.”