All good things must come to an end. So did Build 2018 as Microsoft continues its quest to bring machine learning to every application
Final day started with Demystifying Machine and Deep Learning for Developers with explainer-in-chief and Scott-Hanselman-of-AI, Seth Juarez. It was an interest and captivating talk in which Seth and Chris Lauren walked through the math of machine learning, and explained basics behind machine learning and deep learning, with (almost) all the things one need to build an image classifier for one's application.
Next stop was Matt Winker's session on What's new with Azure Machine Learning where he spoke about updates from Azure Machine Learning to let you manage the end to end lifecycle of Machine Learning Development. The updates made it easy to scale to the cloud, deploy through containers to the cloud and the edge, use great frameworks like Tensorflow, PyTorch, Scikit, and more with a great monitoring and management experience. Matt spoke about next wave of ML capabilities to make it faster to build models, even easier to scale, more deployment options, and how you can easily use Azure Machine Learning to build and deploy models for any of your applications.
After that I checked out Noelle LaCharite @NoelleLaCharite of Microsoft Cognitive Services session on 10 Things Developers Need to Know about Building Intelligent Apps which will be part of every software workload in the not-too-distant future. Noelle spoke about how Developers can use Cognitive Services to infuse web and mobile applications with intelligence, allowing developers to build smarter web and mobile apps, regardless of the level of expertise in Machine Learning.
The last session of the day was with no other than Mark Russinovich (@markrussinovich) who explains how blockchains work, including how they use hashes, transactions, blocks and proof-of-work consensus algorithms to build distributed ledgers. He presented the capabilities of some of the most common blockchain networks, then describes how the COCO Framework addresses their limitations to make blockchains suitable for a wide variety of business consortiums. He even launched Mark Coin!
Till Next Time!
