Skip to main content
Alyx is an AI-powered agent designed to help AI engineers build and improve their applications. Alyx is available throughout Arize AX—depending on where you are in the platform, Alyx has different context and exposes different skills so the help you get is always relevant to that surface.

​
Alyx by surface

Alyx adapts to the surface you’re on and exposes different skills:
SurfaceWhere to find AlyxSkills available
Alyx in the trace slideoverOpen Alyx from a trace detail viewTrace troubleshooting, span analysis, annotations, build evals
Prompt PlaygroundAlyx chat in the PlaygroundOptimize prompts, build evals, run experiments
Eval Builder / Task BuilderAlyx on the eval or task builder pageBuild custom evals, configure tasks
Traces table search barAI Search in the filter barNatural language to filter syntax
Traces pageAlyx on the main Traces pageMulti-trace analysis, pattern discovery
Datasets / ExperimentsAlyx on Datasets or ExperimentsAnalyze experiments, manage datasets

​
Cross-Cutting Skills

These skills are available across multiple surfaces:
SkillDescription
ArizeQL GeneratorTranslates natural language or existing code (SQL, Python) into AQL for custom metrics.
Optimize PromptsOptimizes prompts for better quality or specific issues.

​
ML & CV Models

Alyx supports embedding summarization (cluster summary) for ML and computer vision embedding views—select points or clusters on the embedding page and use “Summarize with Alyx” to get a summary of the selection. See Alyx: Embedding Summarization.

​
Use your own integrations with Alyx

You can use your own LLM integrations with Alyx. Supported today: OpenAI models, Anthropic (via AWS Bedrock and Vertex AI). More are coming soon. Add and manage integrations in Settings → Account Settings → Integrations. For setup and reference, see AI provider integrations.
Add and manage LLM integrations for Alyx in Settings

​
Adding context to Alyx

You can give Alyx explicit context so it focuses on the right data and answers with that in mind. There are three ways to add context:
  • Highlight — Select text anywhere on the page (e.g. a span attribute, an error message, or a prompt snippet), then press ⌘ + L (macOS) or Ctrl + L (Windows/Linux). Alyx opens (or stays open) and the selected text is added to your message as context. If nothing is selected, the shortcut just opens or closes Alyx.
  • @ (mention) — In the Alyx input, type @ to open a menu. You can mention a dataset, experiment, project, or span (depending on the surface). Alyx receives the IDs for what you mention so it can scope the conversation to that data.
  • Type it — You can include context in your message. For example, start with “Additional context:” and add lines of text, or reference a trace, dataset, or experiment by ID in your question. Alyx parses these references and uses them to scope its actions.
How Alyx uses context — Whatever you add (highlighted text, @ mentions, or typed references) is sent with your message as input context. Alyx uses it to scope the conversation: e.g. focus on that trace or span, use that dataset or experiment, or use the highlighted text to answer your question. That way you get answers about the data you care about without re-explaining it every time.
Adding context to Alyx: highlight text, use @ mentions, or type references in your message

​
Alyx Settings

  • macOS: ⌘ + L
  • Windows / Linux: Ctrl + L
Edit your hotkey via the Settings modal, accessible through the menu in the top right corner of the Alyx chat interface. More settings will be added here over time.
Alyx Settings modal—edit hotkey and other preferences

​
Data Privacy

Arize’s Alyx is built on Azure OpenAI because of its built-in security and compliance features. This ensures that customer data is protected and not directly exposed to third-party providers. Here’s how it works:
  • Data Processing: Azure acts as the data processor for the prompts and outputs sent to and generated by Alyx. The models are stateless, meaning no prompts or outputs are stored in the model.
  • No Data Sharing or Model Improvement:
    • Inputs and outputs of Alyx are NOT used to improve OpenAI models.
    • They are NOT used to improve any Microsoft or third-party products or services.
    • They are NOT used for automatically improving Azure OpenAI models.
  • Full Control by Microsoft: The Azure OpenAI Service is fully controlled by Microsoft, and the OpenAI models are hosted in Microsoft’s Azure environment. The service does NOT interact with any other OpenAI-operated services, such as ChatGPT or the OpenAI API.
  • Security & Compliance: Azure OpenAI ensures that we meet industry-standard security and compliance measures, protecting your data throughout the process.
For a detailed breakdown of the data flow and additional privacy measures, refer to the diagram below:
To read more about Azure Data privacy, see their documentation here. If you have further questions or need more clarification on how your data is managed, feel free to contact our team at support@arize.com.

​
Third-Party Integrations

Alyx includes a support skill designed to help answer user questions. When you ask a support-related question, the question itself is sent to a third-party service, RunLLM, for processing. It’s important to note that:
  • Limited Data Sharing: Only the specific question you ask is shared with RunLLM. No additional model information or user data is shared beyond the question itself.
  • User Control: You retain control over your interaction with this skill. If at any time you wish to modify or revoke your consent to share support questions with RunLLM, please contact us at support@arize.com
  • Disclaimer Acknowledgement: Before using the support skill, users must acknowledge a one-time disclaimer outlining the involvement of RunLLM.