AI Assistant

The Servyx AI Assistant lets you ask questions about your infrastructure in natural language and get instant, context-aware answers.

How It Works

The AI assistant has access to your full infrastructure context -- every instance, volume, database, load balancer, cost record, and optimization finding across all your connected accounts and clusters. When you ask a question, it analyzes your actual data to give you a specific, actionable answer.

This is not a generic chatbot. It knows your infrastructure.

Getting Started

You can access the AI assistant from your dashboard. Type your question in plain language and the assistant will respond with relevant data, analysis, and recommendations.

Example Questions

Here are some things you can ask:

Cost Analysis

  • "What is my biggest cost driver this month?"
  • "How much am I spending on stopped instances?"
  • "Which account has the highest cost growth?"
  • "What will my AWS bill be next month?"

Resource Investigation

  • "Which EC2 instances have been idle for more than a week?"
  • "Do I have any orphaned EBS volumes?"
  • "What is the average CPU utilization across my production instances?"
  • "How many instances are running without any tags?"

Optimization

  • "What are my top 5 savings opportunities?"
  • "Which instances should I right-size?"
  • "Am I using my Reserved Instances effectively?"
  • "How much could I save by scheduling my dev environments?"

Kubernetes

  • "Which pods are over-provisioned?"
  • "What is the resource utilization on my cluster nodes?"
  • "How much headroom do I have before I need to scale?"

Comparisons

  • "Compare my production and staging account costs"
  • "Which region is the most expensive?"
  • "How has my spend changed since last month?"

What Context Does the Assistant Have

The assistant can reference:

  • All connected AWS accounts -- instances, volumes, databases, load balancers, and their configurations
  • Performance metrics -- CPU utilization, network throughput, disk I/O from CloudWatch
  • Cost data -- Monthly spend by service, cost trends, projections
  • Optimization findings -- All active recommendations and their estimated savings
  • Kubernetes data -- Cluster resources, pod utilization, over-provisioning findings

Language Support

The AI assistant supports multiple languages. You can ask questions in English, Portuguese, Spanish, French, German, and many other languages. The assistant will respond in the same language you use.

Tips for Better Answers

  • Be specific -- "How much am I spending on m5.xlarge instances?" will give a better answer than "Tell me about my costs"
  • Name your accounts -- If you give your accounts descriptive names, you can reference them naturally: "What's the most expensive service in my production account?"
  • Ask follow-up questions -- The assistant remembers the conversation context, so you can drill deeper: "Which ones are in us-east-1?" after asking about idle instances
Previous← Account Deep DiveNextHow Savings Are Calculated →