As companies embed more and more cloud infrastructure and SaaS within their products and internal applications, the question of how to effectively manage those cloud assets becomes increasingly complex.

CloudQuery’s new embedded AI Assistant helps teams understand their complex cloud infrastructure, faster. It uses LLMs and understands your unique schema, generates SQL that can help you answer any questions you have, and gives you insight into why it generated the SQL that it did.

With CloudQuery's AI Assistant, answering questions about your cloud infrastructure is easier than ever: from finding abandoned resources, to security risks, to resources that violate your compliance commitments. Simply type a question, let an LLM generate SQL for you, run that SQL right in your browser, and iterate.

Try out AI Assistant SQL Editor today → Sign up.

Cloud infrastructure is built for a heterogeneous set of users, each with their own unique requirements. Providers like AWS, GCP, and Azure have done their best to streamline how this infrastructure is configured and managed, but still: cloud infrastructure is complex.

Just look at a Terraform provider for a simple object like an S3 bucket to see how this complexity manifests itself: policies, grants, IAM, and CORS rules are just the appetizer. Typical internal applications require at least four to five different cloud objects, each with hundreds of possible configuration settings.

Now multiply those resources by the hundreds or thousands of use cases that large companies support, and the reality that many large companies are multi-cloud, and you end up with a mired web of cloud assets and configurations.

How can security, compliance, and cost teams navigate this maze of cloud infrastructure, and find exactly what they’re looking for? In the past, that typically has been wading through the UIs of multiple tools, writing SQL by hand, or exporting data and wrangling spreadsheets.

CloudQuery was built to solve this problem at its core. It connects directly to your cloud providers and SaaS systems, extracts all of your configuration and resource data, and loads it into a data store of your choice.

The result: a single, queryable source of truth for your cloud. Instead of hopping between dashboards or APIs, teams can ask any question of their infrastructure using standard SQL.

Want to find every security group that allows 0.0.0.0/0 ingress? Or all the resources that haven’t been tagged with an owner? Or which instances are costing you the most across AWS and GCP? With CloudQuery, it’s all in one place.

The embedded AI assistant takes this one step further. Instead of requiring SQL fluency, anyone on your team can simply ask a question in plain English and get an accurate, runnable query in return: tuned to your schema.

The assistant not only generates SQL, but also shows its reasoning: why it joined specific tables, why it filtered by certain columns, and what assumptions it made. This makes it easy to trust the results, iterate quickly, and even learn SQL along the way.

In other words: it’s SQL with training wheels for non-analysts, and a productivity multiplier for experts, all fine-tuned to your unique cloud infrastructure data.

Cloud infrastructure isn’t getting simpler. But with a unified cloud asset inventory and CloudQuery’s embedded AI assistant, you can interrogate your cloud with natural language, get instant answers, and move faster on cost, compliance, and security initiatives.

Try it yourself: sign up here.

If you want to use CloudQuery’s AI assistant with the CLI, get started with by typing cloudquery init or check out our MCP server in IDEs (like Claude) or ChatGPT.