Eric Anderson (@ericmander) is joined by Rajat Monga (@rajatmonga), a co-creator of TensorFlow. Originally developed by the Google Brain team, TensorFlow is now one of the most popular open-source libraries for machine learning. The team at TensorFlow seek to “democratize” the world of AI as we know it, and by all accounts, they are succeeding. Listen to today’s episode to get inside one of the largest and most exciting open-source projects of the decade.
In this episode we discuss:
- How TensorFlow compares to other open-source projects at Google
- Taking bets on launch day numbers
- Balancing the demands of different kinds of TensorFlow users
- Lessons from Keras and PyTorch
Links:
- TensorFlow
- Keras
- PyTorch
- Kafka
- Kubernetes
- MapReduce: Simplified Data Processing on Large Clusters
- Bigtable: A Distributed Storage System for Structured Data
People mentioned:
- Jeff Dean (@JeffDean)
- Andrew Ng (@AndrewYNg)
- François Chollet (@fchollet)