Beyond Off-the-Shelf Consensus
Dr. Rebecca Bilbro is a data scientist, Python and Go programmer, teacher, speaker, and O'Reilly author. As co-founder and CTO of Rotational Labs, an intelligent distributed systems company, she specializes in machine learning optimization and API development in distributed data systems.
She is also an active contributor to open source software and is the creator and maintainer of the popular Yellowbrick library, an open source Python package that hooks into the popular scikit-Learn API to support visual feature analysis, model selection, and hyperparameter tuning for data scientists and machine learning practitioners.
There have never been more commercial tools available for building distributed data apps — from cloud hosting services, to cloud-native databases, to cloud-based analytics platforms. So why is it still so hard to make a successful app with a global user base?
One of the toughest challenges cloud offerings take on is the problem of consensus, abstracting away most of the complexity. That's no small feat, given that this is a hard enough problem that people spend years getting a PhD just to understand it! Unfortunately, while buying off-the-shelf cloud services can accelerate the path to an MVP, it also makes optimization tough. How will we scale during a period of rapid user growth? How do we do I18n and l10n or guarantee a good UX for users on the other side of the world? How do we prevent replication that might get us into legal trouble?
In this talk, we'll consider several case studies of global apps (both successful and otherwise!), talk about the limitations of off-the-shelf consensus, and consider a future where everyday developers can use open source tools to build distributed data apps that are easier to reason about, maintain, and tune.
- 2022 April 8 15:50 PDT
- 20 min
- Silicon Valley 2022