Skip to main content

Documentation Index

Fetch the complete documentation index at: https://hydroxai.mintlify.app/llms.txt

Use this file to discover all available pages before exploring further.

Know Your AI is a security and evaluation platform built to help teams understand, test, and improve AI systems end-to-end. Unlike tools that only surface metrics, Know Your AI helps you plan evaluation workflows, execute them at scale, and deliver clear, actionable findings from datasets, model runs, and human review. This site contains the product overview, quickstart guides, and feature documentation to get your team up and running.

Get started

Create your first workspace, connect a product, and run an offline evaluation.

What is Know Your AI?

Know Your AI is an evaluation and monitoring platform for AI applications. It centralizes:
  • Workspaces for teams and projects
  • Datasets from the Prompt Market and your own uploads
  • Evaluations to score quality, safety, and compliance
  • Monitoring dashboards and alerts
  • Evidence trails for audits and reviews

Who is it for?

  • Product teams validating new model releases
  • ML/AI engineers testing prompts and model variants
  • Risk & compliance teams tracking policies
  • QA and support teams monitoring regressions

Core concepts at a glance

Product

Explore the platform modules and how they work together.

Workspace

Understand project boundaries, roles, and access controls.

Evaluation

Learn how we score quality, safety, and policy compliance.

Monitoring

Track live performance with dashboards and alerts.

Typical workflow

1

Select a dataset

Choose a dataset from the Prompt Market that matches your risk profile.
2

Connect your product

Link your product API so we can run offline evaluations.
3

Run and review

Use LLM-as-judge and human review to surface issues and risks.
4

Monitor in production

Track drift and quality in real time with alerts and reports.

Where to go next

Get started

Launch your first evaluation.

Datasets

Browse curated prompt datasets and attack methods.

Monitoring

Stay on top of model health.