Takeaways from the Google Cloud Next Conference
Last month, Huy, Emily, and Kelsey from our engineering team attended the Google Cloud Next conference in San Francisco. Their themes, takeaways, and tips are below (as well as pictures of Prog, obviously).
Q: What is something unexpected you learned at Google Cloud Next?
Huy: (Not so) fun fact: 95% of principals use less than 3% of their permissions! While it might be tempting to pile on permissions when you’re not sure what you need, keep in mind that users and service accounts are highly susceptible points of attack. Ensuring that your principals are endowed with only restricted access rights is pivotal to minimizing security breaches.
Emily: I was surprised by just how heavily Google emphasized Generative AI as a feature they are rolling out across a number of products. They really drove home the point that “GenAI is here, and it’s going to change everything, and you should use it.” We are still figuring out what role GenAI will play for us at CTA as we take the security of our data very seriously. However, it is clear it will be used widely in both the private and public sectors, and that’s a shift we will need to stay informed about. And some GenAI tools, like natural-language-to-SQL, could provide real value to our users, and that’s exciting!
Q: What is coming on the horizon that you were excited to share with your team?
Huy: Surprisingly, Google Meet with Duet AI! I’m a big multi-tasker for better or worse, so during meetings, I have a hard time not looking away for a minute to work on a small task or write a bit of code. Sometimes this can cause me to miss a key point/action item. Google is rolling out new features for Meet with Duet AI, including generating action items, mid-meeting summaries, and a chatbot that can talk about details discussed during an ongoing call.
Kelsey: I’m excited to see how tools utilizing natural language processing, like Duet AI and BigQuery Studio, may be able to fill any technical gaps for folks that we partner with so that they can get the most out of PAD and reduce the need for technical expertise in order to be productive. The idea of being able to use natural language directly within BQ to analyze data is exciting, and I can see that being a major time saver — even for folks who use SQL regularly but may not know the specific syntax they need to accomplish a task.
Q: What are some tips that you learned at the conference that people can use right now?
Huy: You can use the Security Command Center in Google Cloud to scan for vulnerabilities in your organization and detect misconfigurations that may be weakening your security! There’s a free tier to try, while the premium tier offers enhanced scanning, threat detection, and more.
Kelsey: Utilize labels in ETL/ELT jobs to organize resources and track costs throughout your project. Information about labels is forwarded to the billing system, so you can break down billed charges by label. This way resource usage can be easily attributed to specific teams, projects, partners, etc.
Emily: As Kelsey mentioned, there are a lot of tools that are already available for tracking costs, both in terms of analysis (running queries) and storage (maintaining data in BigQuery). The INFORMATION_SCHEMA views in BigQuery make it easy to estimate storage costs by looking at the size of tables, and you can also configure GCP to create a BigQuery table with logs from when users run queries, which makes it easy to track how much each user or project is spending on queries, and how much each individual query costs. How neat is that!
The team with Prog at the conference, catching some downtime and participating in trainings.