# arize.com

> AI-optimized mirror of arize.com containing 50 pages totalling 27,866 words of clean markdown content, structured data, and semantic HTML. Original source: https://arize.com. Last updated: 2026-07-20T14:24:13.129Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [Agent Observability, Evaluation & Improvement Platform | Arize AI](/content/site-root.html): Continuously improve AI agents with agent observability, evaluation, tracing, and experimentation. (711 words)

## Articles & Blog Posts

- [blog/managing-and-monitoring-your-open-source-llm-applications.html](/content/blog/managing-and-monitoring-your-open-source-llm-applications.html) (2,445 words)
- [Refresh an API key - Arize AX Docs](/content/docs/api-reference/api-keys/refresh-an-api-key/index.html): Atomically revoke an existing API key and issue a replacement with the same metadata (name, description, and key type). The old key is invalidated and the new key is activated in a single transaction — there is no window where neither key is valid. The full new key value (key) is only returned once in the response. Store it securely. Authorization: - User keys: the creator or an account admin may refresh the key. Requires the developer user permission flag. Returns 403 when this flag is absent. - Service keys: space admins (and higher) may refresh any service key in their space. Non-admins require the SERVICE_KEY_CREATE permission and must be the creator of the key. Expiry behaviour: expires_at is required when the existing key has an expiry — omitting it would extend the key's lifetime to unbounded and is rejected with 422. For unbounded existing keys, expires_at may be omitted (the replacement is also unbounded) or supplied to add a specific expiry. The value must not be later than the existing key's expiry; to issue a key with a longer lifetime, use POST /v2/api-keys. Grace period: Supply grace_period_seconds in the request body to keep the old key valid for that many seconds after the refresh. If not supplied, the old key is revoked immediately. This endpoint is in beta, read more here. (352 words)
- [VertexAI - Arize AX Docs](/content/docs/ax/security-and-settings/integrations-playground/vertexai/index.html): Integrate with the VertexAI API as an AI Provider to get access to Google models (515 words)
- [Using Evaluators - Phoenix](/content/docs/phoenix/datasets-and-experiments/how-to-experiments/using-evaluators/index.html) (1,101 words)
- [Improving Structured Output Generation with Prompt Learning - Arize AX Docs](/content/docs/ax/cookbooks/improve/improving-structured-output-generation-with-prompt-learning.html): In this guide we use Prompt Learning to improve the accuracy of GPT-4o-mini on structured output generation. (1,039 words)
- [Iterative Evaluation & Experimentation Workflow (TypeScript) - Phoenix](/content/docs/phoenix/cookbook/ai-engineering-workflows/iterative-evaluation-and-experimentation-workflow-typescript.html): Phoenix Tracing, Evaluating, and Experimentation Walkthrough (1,607 words)
- [Get an AI integration - Arize AX Docs](/content/docs/api-reference/ai-integrations/get-an-ai-integration/index.html): Get a specific AI integration by its ID. This endpoint is in alpha, read more here. (441 words)
- [List role bindings - Arize AX Docs](/content/docs/api-reference/role-bindings/list-role-bindings/index.html): List role bindings for the authenticated user's account, filtered by resource type. Results are paginated; use limit and cursor for subsequent pages. The resource_type query parameter is required and must be one of SPACE or PROJECT. All bindings in the account are visible to any authenticated account member. Use user_id to narrow to a specific user. This endpoint is in beta, read more here. (350 words)
- [AI Search: Your AI Debugger - Arize AI](/content/resource/ai-search-your-ai-debugger/index.html): Arize AI Agent Search helps engineers quickly identify patterns, uncover problematic traces, and debug applications with ease. (437 words)
- [Arize:Observe - Fireside Chat with Anthony Goldbloom - Arize AI](/content/resource/fireside-chat-with-anthony-goldbloom/index.html): Detecting, diagnosing, and resolving ML model performance can be difficult for even the most sophisticated ML engineers. (259 words)
- [List annotation configs - Arize AX Docs](/content/docs/api-reference/annotation-configs/list-annotation-configs/index.html): List annotation configs the user has access to. This endpoint is in beta, read more here. (406 words)
- [Get an organization - Arize AX Docs](/content/docs/api-reference/organizations/get-an-organization/index.html): Get a specific organization by its ID. This endpoint is in beta, read more here. (448 words)
- [04.18.2025: Tracing for MCP client server applications - Phoenix](/content/docs/phoenix/release-notes/04-2025/04-18-2025-tracing-for-mcp-client-server-applications.html): Available in Phoenix 8.26+ (199 words)
- [Run Experiments with LLM as a Judge - Arize AX Docs](/content/docs/ax/develop/tutorial/run-experiments-with-llm-judge/index.html): Learn how to define and run experiments to systematically evaluate your AI application using LLM as a Judge evaluators for subjective quality assessments. (650 words)
- [NVIDIA GTC | Arize - Arize AI](/content/nvidia-gtc-2025/index.html): Meet the Arize team at NVIDIA GTC 2025. Learn about AI Observability and LLM Unified Observability and Evaluation Platform for AI (504 words)
- [Agent Workflow Patterns - Phoenix](/content/docs/phoenix/cookbook/agent-workflow-patterns/index.html): Workflows are the backbone of many successful LLM applications. They define how language models interact with tools, data, and users—often through a sequence of clearly orchestrated steps. Unlike fully autonomous agents, workflows offer structure and predictability, making them a practical choice for many real-world tasks. (411 words)
- [Resource - Arize AX Docs](/content/docs/ax/concepts/otel-openinference/resource/index.html): The immutable metadata that describes who is producing telemetry — service name, environment, host, and the Arize AX attributes that route spans to your project. (615 words)
- [Prompt flow - Phoenix](/content/docs/phoenix/integrations/platforms/prompt-flow/index.html): PromptFlow is a framework for designing, orchestrating, testing, and monitoring end-to-end LLM prompt workflows with built-in versioning and analytics (30 words)
- [List resource restrictions the caller is permitted to manage. - Arize AX Docs](/content/docs/api-reference/resource-restrictions/list-resource-restrictions-the-caller-is-permitted-to-manage.html): List active resource restrictions the authenticated user is permitted to manage. A restriction is returned only if the caller can manage it — i.e. an account/org admin (via admin escalation), a holder of the PROJECT_RESTRICT permission in the project's space, or a holder of PROJECT_RESTRICT granted directly on the project. Results are paginated; use limit and cursor for subsequent pages. Because entries are authorization-filtered after a page is read, a page may contain fewer items than limit (or be empty) while has_more is still true. Clients MUST keep paging until has_more is false — do not stop on an empty page. Use the optional resource_type query param to filter to a single resource type. When omitted, PROJECT restrictions are returned (currently the only supported type). This endpoint is in beta, read more here. (399 words)
- [False Negative - Arize AI](/content/glossary/false-negative/index.html) (43 words)
- [Phoenix 13.0 - Phoenix](/content/docs/phoenix/release-notes/02-2026/02-14-2026-phoenix-13-0/index.html): Dataset Evaluators, custom model providers, OpenAI Responses API support, and major Playground and experiment UX improvements. (926 words)
- [03.19.2025: Access to new integrations in projects - Phoenix](/content/docs/phoenix/release-notes/03-2025/03-19-2025-access-to-new-integrations-in-projects/index.html): Available in Phoenix 8.15+ (42 words)
- [Release Notes - Phoenix](/content/docs/phoenix/release-notes/03-2026/03-30-2026-delete-prompts-api/index.html) (136 words)
- [Datasaur - Label Data with Premiere Quality and Productivity Using Datasaur - Arize AI](/content/resource/datasaur-label-data-with-premiere-quality-and-productivity-using-datasaur.html) (34 words)
- [In Defense of Central ML - On Demand - Arize AI](/content/resource/in-defense-of-central-ml/on-demand/index.html) (62 words)
- [Arize AI Un/Summit - Model Improvement - Arize AI](/content/resource/arize-ai-un-summit-model-improvement/index.html): This chat is a part of Arize AI's ML Observability Un/Summit 2020 Panelists: John Trenkle - Principal Machine Learning Engineer @ TubiChristopher Brown- Chief Data Scientist @ ADARA, Inc. (61 words)
- [AI Research - Arize AX Docs](/content/docs/ax/cookbooks/ai-research/index.html): Collection of advanced experiments and benchmarks in LLM evaluation, instrumentation, and agent systems (173 words)
- [Release Notes - Phoenix](/content/docs/phoenix/release-notes/04-2026/04-01-2026-get-traces-secrets-api-and-python-314/index.html) (404 words)
- [GraphQL API - Arize AX Docs](/content/docs/ax/graphql-reference/index.html) (386 words)
- [Sarah's Test Page - Arize AI](/content/resource/sarahs-test-page/index.html) (69 words)
- [Chunking strategy - Arize AI](/content/glossary/chunking-strategy/index.html) (75 words)
- [09.26.2024: Authentication & RBAC - Phoenix](/content/docs/phoenix/release-notes/2024/09-26-2024-authentication-and-rbac/index.html): Available in Phoenix 5.0+ (184 words)
- [Hugging Face Smolagents - Phoenix](/content/docs/phoenix/integrations/python/hugging-face-smolagents/index.html): Hugging Face smolagents is a minimalist Python library for building powerful AI agents with simple abstractions, tool integrations, and flexible LLM support (35 words)
- [Model Performance - Arize AI](/content/glossary/model-performance/index.html) (86 words)
- [Prompt injection - Arize AI](/content/glossary/prompt-injection/index.html) (78 words)
- [The agent is the user now: lessons from the founder of WorkOS - Arize AI](/content/blog/ai-engineer-agents-workos-identity-permissions-evals.html): WorkOS founder Michael Grinich says the next era of AI engineering will not be defined by prompts alone. (1,696 words, Jul 8, 2026)
- [How to evaluate AI agents, avoid reward hacking, and build better specs - Arize AI](/content/blog/how-to-evaluate-ai-agents-and-build-better-specs/index.html): Agent evals are repeatable tests that score whether AI agents completed a task correctly. Learn how to design rubrics, test suites, and trace-based evals that catch failures and prevent reward hacking. (1,702 words, Jul 2, 2026)
- [100 AI Agents Per Employee: The Enterprise Governance Gap - Arize AI](/content/blog/100-ai-agents-per-employee-governance-gap/index.html): NVIDIA GTC signaled massive AI agent adoption. But most enterprises can’t see what agents do in production. Learn how to close the governance gap. (1,108 words, Mar 22, 2026)
- [Arize Release Notes: Labeling Queues, Expand/Collapse Rows in Trace Table - Arize AI](/content/blog/arize-releases-labeling-queues-expand-collapse-rows-in-trace-table.html): New features and enhancements from Arize including labeling queues, expand/collapse rows, and the latest content. (189 words, Mar 4, 2025)
- [Arize Release Notes: Prompt Hub, Managed Code Evaluators and More - Arize AI](/content/blog/arize-release-notes-prompt-hub-managed-code-evaluators-and-more.html): New features and enhancements from Arize including the prompt hub: a central repository for managing, iterating & deploying prompt templates. (482 words, Dec 19, 2024)
- [AI Agent Workflows and Architectures Masterclass - Arize AI](/content/blog/ai-agent-workflows-and-architectures/index.html): Insights on agent workflows and the future of agent-driven architectures from Jerry Liu, Co-Founder and CEO at LlamaIndex & Jason Lopatecki. (946 words, Dec 4, 2024)
- [Arize Release Notes: Sep 5, 2024 - Arize AI](/content/blog/arize-release-notes-sep-5-2024/index.html): Regular updates on new releases, enhancements, and changes. (166 words, Sep 5, 2024)
- [Introducing Arize Copilot - Arize AI](/content/blog/introducing-arize-ai-copilot/index.html): If you used Microsoft Office in the early days, you probably remember Clippy. Clippy was an animated paper clip and go-to assistant for all things Microsoft Office. It provided users with tips, help, and shortcuts for using Microsoft Office applications. Well, it got us thinking…what if there was a Clippy but AI…for AI? Introducing Arize... (1,317 words, Jul 11, 2024)
- [Hugging Face + Arize: Partnership and Code Example - Arize AI](/content/blog/arize-hugging-face/index.html): Our partnership with Hugging Face helps organizations train unstructured models and monitor and troubleshoot those models in production. (2,301 words, Dec 22, 2022)
- [Introducing the Arize Trust Center and Security Periodic Table - Arize AI](/content/blog/introducing-the-arize-trust-center-and-security-periodic-table.html): Since joining Arize as Chief Information Security Officer earlier this year, it has been inspiring to see the tangible benefits that Arize’s ML observability platform brings every day to both free and enterprise users. Despite only being in the early stages of this industry, Arize is already trusted by its customers to process hundreds of... (466 words, Jun 1, 2022)
- [The Chronicles of AI Ethics: The Man, The Machine, And The Black Box - Arize AI](/content/blog/the-chronicles-of-ai-ethics-the-man-the-machine-and-the-black-box.html): As we give our technology more and more responsibility, the importance of making ethically charged decisions in our model building is amplified exponentially. It all boils down to really understanding our creations. If we don’t know what is happening in the black box, we can’t fix its mistakes to make a better model and a better world. (1,141 words, Mar 12, 2021)

## Listings & Categories

- [Posts by Xander Song — Arize](/content/author/xander-song/index.html) (13 words)
- [AI In the Enterprise Archives - Arize AI](/content/blog/category/ai-in-the-enterprise/index.html) (601 words)
- [Posts by Natalia Skaczkowska-Drabczyk — Arize](/content/author/natalia-skaczkowska-drabczyk/index.html) (25 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/robots.txt): Crawler directives