We’ve seen in previous posts what Observability as Code (OaC) is and how it’s now essential for managing observability resources at scale (Self-Service Observability: How To Scale Observability Adoption Through Self-Service, How To Build a Self-Service Observability Practice with Splunk Observability Cloud). We also saw how to build out an OpenTofu/Terraform repository to successfully start the OaC practice (Splunk Observability as Code: From Zero to Dashboard). But what if it could be even easier?
What is the Splunk Observability Cloud AI Assistant?
If you haven’t seen our series on Splunk Observability Cloud’s AI Assistant in Action (find the first post here), we’ll give you a bit of background. The AI Assistant is backed by Agentic AI and is built directly into Splunk Observability Cloud. It helps you quickly and easily tap into the health of your applications and infrastructure and can even do cool things like generate Terraform code for your observability resources. Instead of manually writing HCL syntax from scratch, you can describe what you want to build in natural language and get production ready Terraform code in seconds.
Creating a Dashboard using the Splunk Observability Cloud AI Assistant
Let’s walk through the simple steps of building a dashboard using the Splunk AI Assistant. Let’s say we want a dashboard for monitoring our Kubernetes cluster’s health (side note: Splunk Observability Cloud comes with out-of-the-box Kubernetes navigators so we can monitor Kubernetes environments with no manual configuration, but stick with us here).
Step 1: access the AI Assistant
Navigate to Splunk Observability Cloud and look for the AI Assistant icon. Select it to open the assistant:
Step 2: describe your dashboard
In the AI Assistant, describe what you want in natural language:
The assistant will generate Terraform code, which you can then plug into your Observability as Code repository:
Don’t yet have an OaC repo template that communicates with Splunk Observability Cloud? Here you go: http://cs.co/self-service-observability
Building Charts with the AI Assistant
We can then populate our Dashboard with charts by generating chart configurations via Terraform using the AI Assistant. In our prompt, we can specify parameters like the dashboard we’d like the chart to live on, the metrics we’d like to monitor, the time range we’d like to focus on, the way we would like our data grouped, etc.:
We can then take that Terraform generated by the Assistant, and once again plug that into our Observability as Code repository.
Once we’ve deployed those new resources, along with any others we wish to create, we can use them to monitor our application from within Splunk Observability Cloud:
Best Practices
When using the AI Assistant for Terraform generation:
- Be specific and include as much detail as possible – metric names, filters, aggregation methods, time range, etc.
- Iterate. Start with a basic dashboard and chart, then refine to add more specific details
- Always review generated code to validate configurations.
- Dashboard groups require a dashboard_group resource, so make sure to create one of those first.
- Ask the Assistant to use variables and parameterize common values like namespaces or clusters.
Next steps
Ready to try it for yourself? Head over to Splunk Observability Cloud, open the AI Assistant, and start building. Don’t yet have Splunk Observability Cloud? Try it free for 14 days.
Resources
- Prompt guide and library for AI Assistant in Observability Cloud
- Splunk Terraform Provider documentation
- Connect your cloud services using Splunk Terraform
Never miss a new post. Check out this short guide on how to subscribe to the blog and get updates.
