Managed Workflow for Apache Airflow (MWAA): Latest and Greatest!
With Jen Dunham. Choosing between MWAA deployment models and exploring recent platform developments.
Ideas and practical lessons from building products, cloud infrastructure, and developer tools.
Talks & videos · Technical articles · Research
With Jen Dunham. Choosing between MWAA deployment models and exploring recent platform developments.
Practical ways to reduce monitoring costs while retaining useful signals. Session page includes the talk and transcript.
Deep dive into AKS cluster troubleshooting techniques covering node saturation metrics for performance optimization, leveraging Kubernetes events as real-time cluster signals, and fine-tuning resource allocation with cluster autoscaler metrics.
Deep dive into the latest advancements from AKS for improved troubleshooting including new AI-based features. This session covers the AKS monitoring and troubleshooting stack, challenges with monitoring, best practices, and includes live demos of Retina for network observability. Co-presented with Pavneet Ahluwalia and Neha Aggrawal.
Co-authored guide to running custom Python and shell tasks in serverless Airflow workflows.
Introduction to HolmesGPT, an open-source agentic AI framework for root cause analysis in cloud native environments. Co-authored with the Robusta.dev team, this post explores how AI agents can revolutionize Kubernetes troubleshooting through extensible toolsets, natural language prompts, and intelligent diagnostics.
Announcing native Grafana dashboards in Azure Portal for AKS clusters at no additional cost. This integration eliminates the complexity of maintaining separate visualization tools while delivering comprehensive cluster observability with Container Insights, Prometheus, and Azure Monitor metrics out-of-the-box.
Introducing the CLI Agent for AKS - an AI-powered command-line experience for troubleshooting, optimizing, and operating AKS clusters. Built on open-source HolmesGPT (CNCF Sandbox project) and the AKS Model Context Protocol server, this human-in-the-loop tool brings intelligent agentic workflows directly to your terminal with a focus on security and transparency.
Monitoring the health and resource utilization of the AKS control plane.
Routing AKS platform metrics to the monitoring tools and destinations you use.
Aritra Ghosh and C. Chandra Sekhar.
Aritra Ghosh and P. Gudipati.
Aritra Ghosh and J. Vachery.