Khaga – AI Infrastructure Diagnosis for AWS, GCP, Azure and Kubernetes

Name: Khaga – AI Infrastructure Diagnosis for AWS, GCP, Azure and Kubernetes
Availability: InStock
Author: Gowrishankarhq

by Gowrishankarhq·Mar 5, 2026·1 point·2 comments

Visit Project View on HN

AI Analysis

●●SolidSolve My ProblemDark Horse

Multi-cloud diagnosis in <30s, but infra observability (Datadog, New Relic) already solves this better.

Strengths

•True multi-cloud support (AWS, GCP, Azure, K8s) in single pane of glass, rare integration
•Zero-agent architecture—no deployment friction, just API credentials encrypted with Fernet
•Terraform plan review + CI/CD log parsing + compliance estimates add real breadth

Weaknesses

•Relies entirely on Claude API—no local models or self-hosted option for security-conscious teams
•No evidence of real SRE adoption; landing page is polished but product maturity unclear

Post Description

I built this after getting frustrated with jumping between CloudWatch, kubectl, and 4 other tools every time something broke in prod. You point it at your AWS, GCP, Azure, or Kubernetes setup and it gives you a root cause analysis in plain English — severity, evidence, and fix commands. Also does Terraform plan review, Dockerfile analysis, CI/CD log parsing, and SOC2/ISO27001 compliance estimates. Free to try, no credit card. Would love harsh feedback from people who actually manage infrastructure.