So as you saw in the quick demo, the Cloud Datalab notebook is one single place where you can have documentation, code, results, and visualizations all in one place. And one of the great things about it is, since data scientists use these tools prolifically, there's a lot of existing notebook examples that you can just pull down from the web. And Datalab of course is based on the open source project Jupyter, that's if you've heard of Jupyter notebooks. It's pretty much just a container of Jupyter notebooks that runs on Google Cloud. The main benefit here of course is if you're a data scientist and you're used to running these IPython notebooks locally on your machine, now, you can actually take advantage of running these on a virtual machine. That, as you saw on the demo, has direct access to big query level processing and cloud machine learning APIs, and the power and computational strength of Google Compute Engine. And if you needed to store all that data, you have something called Google Cloud storage that you can immediately write those results to. So it's all the power and flexibility of a cloud notebook with a benefit of really fast connections to a lot of these other big data products on Google Cloud Platform. And if building Machine Learning models is something that interests you, this is definitely one of the largest first steps that you're going to take. As you saw, we invoked a lot of SQL within the cells of Cloud Datalab. And if you get familiar with how Cloud Datalab notebooks are set up and get a little bit more dangerous with Python and R, and invoking the Cloud Machine Learning engine APIs, that's exactly what we're going to be doing in the next specialization which is data engineering. So if that's something that interests you, I highly recommend for you to take that course. >> So as you just saw, Cloud Datalab notebooks are much like the BigQuery web UI. But the addition is that you can add multiple queries and you can add mark down for color commentary. Cloud Datalab is heavily used with Machine Learning APIs like TensorFlow. And generally, you'll find that data scientists will preprocess all the data in BigQuery and then explore and build those ML models inside of Cloud Datalab. Lastly, those notebooks are meant to be shared and presented with your other data scientist peers. And your audience will be able to follow the exact same steps that you've took in your analysis and provide you with helpful feedback.