Intro to Overview of Builder Workloads
1. Intro to Overview of Builder Workloads
Hooray hooray! You’ve made it to the final module of the course. I think this is very exciting (!), both because you’ve come a long way – I mean, you’ve learned about the Snowflake CLI. How cool are you? But also because you now know enough about the building blocks of Snowflake for us to get deeper into some of Snowflake’s workloads. This is extremely useful stuff to know, and I’m hoping that knowing this can help you move forward your career, whether that’s getting a job at a company that uses Snowflake, or becoming more effective at your current role. This third part of the course will differ from the previous two in a few ways – So get ready for a slight change in style. The first difference is that we’ll look at the Snowflake platform through a workload-specific lens. What does that mean? You can think of Snowflake’s platform as being made up of functionality that supports a bunch of different workloads, including: One, Applications. Two, Collaboration. Three, Data Engineering. Four, Data Lake. Five, Unistore (which are our hybrid tables – so they’re part transactional and part analytical) Six, AI / ML. Seven, Data Warehouse, and Eight, Cybersecurity. So when we take a workload-specific approach, we focus on buckets of related features, and tackle each bucket in turn. This is a useful way to look at Snowflake, because it makes it easier for us to create a mental map of what Snowflake does, but it comes at a cost, which is that it can make the parts seem more disconnected than they really are. So throughout, we’ll have to try to mitigate the downsides of this workload-specific lens by reminding ourselves that Snowflake is one product. The features are interconnected, and it’s not the case that one kind of data practitioner is just going to use tools from one workload. Many data scientists use data engineering functionality, many app developers use collaboration functionality, and many data engineers are interested in ML. The second way this part of the course will differ from what’s come before is in the structure – So for each workload, we’ll have two videos. In the first video, I’ll give an overview of what can be done with that workload on Snowflake, and I’ll share code snippets related to each thing we discuss. In the second video, we’ll actually use the product to explore one aspect of that workload in a hands-on way. So if you’re ever feeling like the first video for each workload is too high-level, take heart: We’ll be running code for that workload soon enough. Since this is a course focused on Builders – those who build pipelines, statistical models, ML models, data apps – we’re going to tackle the workloads that I think of as the most “Builder-y.” Those are Data Engineering, AI / ML – which we’ll split into separate videos – and Applications. As for the other workloads, we’ve actually already covered a lot of content from the Data Warehouse workload, and the others – Cybersecurity, Unistore, Collaboration, and Data Lake – are important, but not the focus here. The good news is that it won’t differ in a key way – We’ll still often use the Tasty Bytes food truck data that we’ve been using throughout the course. So some things change, some things stay the same. Tasty Bytes, I will never leave you. Here’s what we’ll cover in the hands-on portions: For DE, we’ll practice ingesting streaming data with Snowpipe. For Snowflake’s GenAI workload, we’ll use the Snowflake Cortex LLM function called “Complete” to query an LLM within Snowflake. For Snowflakes’ ML workload, we’ll use Snowpark ML to create an XGBoost model and make predictions about where a particular food truck will be over time. And for Snowflake’s app workload, we’ll make a Streamlit app that shows us Tasty Bytes’ daily revenue over time by country. Much of what I just said was likely very new, but don’t worry – It will all make sense soon. Our goal for this module isn’t to cover any workload comprehensively – we’ll do more of that in future coursework – but instead for you to develop a foundational mental map of the data engineering, AI, ML, and applications work you can do with Snowflake. We’ll also get some hands-on experience using a key feature from each workload. Let’s get started!2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.