Some of the best technical insights come from sharing what you've built and what you've learned along the way.
That's one of the core ideas behind Unprompted, the blog we launched on Medium earlier this year to give Grafana Labs engineers and our broader OSS community a space to showcase their work. From behind-the-scenes engineering stories and technical deep dives to creative use cases for Grafana (a fishing dashboard, anyone?), it's a place to share ideas and learn from one another.
To give you a sense of what you'll find, we've rounded up a few recent posts covering everything from Grafana's architecture to monitoring your favorite reality TV show.
Behind the scenes at Grafana Labs
These posts explore some of the technical challenges our teams have tackled at Grafana Labs, including the decisions, tradeoffs, and lessons that shaped their work.
Our journey to improve Grafana Assistant dashboarding mode
Making an AI agent faster, more reliable, and less expensive takes a lot more than fine-tuning your prompts. Senior Software Engineer Dominik Broj shares how the team used production telemetry, offline evaluations, and Agent Observability to improve dashboarding mode in Grafana Assistant, reducing error rates by 73%, latency p95 by 50%, and cost by 18%.
How and why we're bringing data sources out of the Grafana monolith
Grafana data sources have historically been closely tied to the Grafana monolith. To make it easier to develop, maintain, and release data source plugins independently of Grafana's core release cycle, the team has been working to decouple them from the monolith. But what does that entail, exactly? Kate Vredenburgh, Senior Manager, Engineering, walks through it all, including the challenges of migrating existing code and how the team used AI agents to help along the way.
How we provision Grafana Play dashboards at scale with Git Sync and Terraform
Grafana Play is our public sandbox where you can explore hundreds of Grafana dashboards and dozens of data sources. It’s a great way to learn about and experiment with Grafana, but keeping all those dashboards consistent across development and production environments takes work. Senior Developer Advocate Marie Cruz shares how the team uses Git Sync (because who doesn’t love dogfooding?) and Terraform to manage dashboards as code and streamline the deployment process.
Observability tutorials and how-tos
From troubleshooting CI/CD pipelines to monitoring Kubernetes workloads and analyzing API gateway logs, these technical how-tos offer practical guidance for tackling common observability challenges.
How to build workload-aware SLO dashboards in Grafana for mixed Kubernetes workloads
A Kubernetes dashboard can look healthy overall, even when there are performance issues with individual workloads. In this post, community member and SRE Sai Joshitha Kathari shows how to build Grafana dashboards that account for different workload types, using Prometheus metrics, SLOs, error-budget burn rates, and alerting to get a more accurate view of reliability.
From gateway logs to actionable insights with Apache APISIX, Loki, and Grafana
API gateway logs contain valuable information about traffic, errors, and latency, but making sense of that data isn't always straightforward. In this tutorial, community member and Apache APISIX technical contributor Yilia Lin explains how to send structured logs from APISIX to Loki, then use LogQL and Grafana dashboards to identify failing routes, investigate latency, and better understand API behavior during incidents.
Troubleshooting CI/CD pipelines (and more) with OpenTelemetry traces and Grafana Cloud
When a CI/CD pipeline slows down, figuring out which steps are responsible often means piecing together timestamps from logs. Fortunately, there’s a better way. Matt Simonsen, Senior Solutions Engineer at Grafana Labs, explains how to instrument a GitHub Actions workflow with OpenTelemetry and Traces Drilldown in Grafana Cloud to make your pipeline investigations easier.
Creative and fun use cases for Grafana
Some of the most interesting Grafana use cases are also the most unexpected. Whether you’re planning your next fishing trip or want to keep closer tabs on your favorite TV show, these posts inspire fun and creative ways to use Grafana.
Gone fishin' with Grafana Cloud: How a dashboard can help you catch more fish
Ever wished you could check all the conditions for a fishing trip in one, centralized place? Thomas Mack, Senior IT Operations Engineer at Grafana Labs, shares how to do just that with a Grafana Cloud dashboard that helps you visualize tides, wind, water temperature, and marine forecasts to decide when it's best to head out on the water.
Tracking 'The Traitors' TV show with SQL expressions in Grafana
What can a reality TV competition teach you about SQL expressions? Probably more than you think. Grafana Labs Senior Developer Advocate Simon Prickett turns his enthusiasm for “The Traitors” into a hands-on tour of SQL expressions in Grafana, combining data from Google Sheets and other sources to visualize the show’s voting patterns, eliminations, and more.
From small talk to shared stories: Building a weather dashboard for distributed teams
For distributed teams, even something as simple as talking about the weather can be a way to connect. This post, written by Yaëlle Chaudy, Senior Manager, Engineering at Grafana Labs, explores how a dashboard in Grafana Cloud can turn everyday weather data into a shared experience, giving teammates across the globe a new way to kick off conversations in their daily syncs.
Find more stories on Unprompted
These are just a few of the stories we're sharing on Unprompted. Whether you're interested in behind-the-scenes engineering stories or simply curious about what others in the community are building, there's plenty more to explore.
Follow Unprompted on Medium for more technical deep dives, engineering lessons, and stories from the Grafana community.
Have an interesting Grafana project or technical story of your own? We're always looking for community contributors and would love to hear from you.


