AI/ML + Grafana: How to Create Intelligent Grafana Apps Leveraging LLMs

on YouTube ↗23m

" summary

Grafana adds an LLM application plugin and embeds LLMs into Scenes apps to power concrete features: a dashboard assistant that generates panel titles/descriptions, a query advisor that semantic‑searches a vector database of thousands of PromQL templates and returns the top five for an LLM to adapt, and flamegraph AI that highlights bottlenecks and suggests fixes. Live demo shows streamed chat, switching to 20 servers, and the LLM flagging a SQL4 spike compared to historical baselines.

AI-written summary · about

" chapters (9)

1. 0:00Introduction
2. 0:25What does the future of observability look like?
3. 1:11How do we enable humans to understand complex systems?
4. 2:41GenAI and ML in observability
5. 3:08How we're using AI and ML at Grafana: Minimize toil with Grafana Sift and Grafana ML
6. 6:02Building GenAI support into Grafana OSS
7. 7:49Features we've built so far: Dashboard assistant, query advisor, Flamegraph AI
8. 11:53What can you build?
9. 13:38Grafana, LLMs, Scenes live demo

" related talks

• 23mAI/ML + open source observabilityYasir Ekinci, Ben Sully · ObservabilityCON 2023
• 30mBuilding Grafana AssistantMat Ryer · bit summit 2025
• 07mLightning Talk: Demystifying Kubernetes Observability with Generative AI and LLMsAsaf Yigal · Observability Day Europe 2024
• 01h 00mWorkshop: Getting Data Ready for AI - With Unified GraphQL &SQL EndpointsAnushrut Gupta · GraphQLConf 2024
• 16mIs that LLM feature any good?Simon P. Couch · posit::conf(2025)
• 48mHands-on guide to monitor your API-driven AI/LLM applicationsJones, Zameer · Apidays India 2025
• 38mQuerying Live Data With LLM AppBobur Umurzokov · Big Data Conference Europe 2023
• 24mDemocratizing data access through LLMPee Tankulrat · XConf Thailand 2023