--- title: AI-103 Develop AI Apps and Agents on Azure Study Cram description: A study guide for the Microsoft AI-103 certification (which replaces AI-102), walking through Azure's Microsoft Foundry platform for building generative AI... canonical: https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: AI-103 Develop AI Apps and Agents on Azure Study Cram | daily.dev og:description: A study guide for the Microsoft AI-103 certification (which replaces AI-102), walking through Azure's Microsoft Foundry platform for building generative AI... og:url: https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5 og:image: https://api.daily.dev/og/posts/fX51Njoo5.png og:image:alt: AI-103 Develop AI Apps and Agents on Azure Study Cram og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # AI-103 Develop AI Apps and Agents on Azure Study Cram **[John Savill's Technical Training](https://daily.dev/sources/john-savill)** · 117 min read · 0 upvotes · 0 comments ## Summary A study guide for the Microsoft AI-103 certification (which replaces AI-102), walking through Azure's Microsoft Foundry platform for building generative AI apps and agents. Covers responsible AI principles, the model catalog (closed vs open weight, serverless vs managed compute), deployment options (global/data zone/batch/provisioned throughput), the model router for automatic model selection, the Responses API vs Chat Completions API, tools (code interpreter, MCP, agent-to-agent, toolboxes, skills), Foundry IQ knowledge sources, evaluation/red-teaming, and guardrails for safety. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the difference between the Foundry responses API and the OpenAI chat completions API in Azure AI Foundry? The responses API is stateful, storing conversation history server-side so you only need to reference a previous response identifier for multi-turn interactions, and it works with both Foundry and OpenAI endpoints. Chat completions, by contrast, requires you to manually resend the full message array (including the system role message) on every call and only works with the OpenAI API, meaning context grows over time and you manage state yourself. _Developers comparing Azure Foundry API options can track these workflow details on daily.dev._ ### What is a model router in Azure AI Foundry and how does it choose which model to use? A model router is a Foundry deployment that sits between an application and multiple underlying models (mini, nano, GPT variants, plus Llama, DeepSeek, Grok, and Anthropic when configured), using a small language model to automatically pick the best option for each request and route to it, optimizing cost without requiring custom logic. Similar behavior powers the auto option in Microsoft 365 Copilot and GitHub Copilot. _Teams weighing multi-model strategies can follow model router developments on daily.dev._ ### What are the deployment throughput options for models in Azure AI Foundry, and how do global versus data zone deployments differ? Azure AI Foundry offers global standard, data zone standard, global batch, data zone batch, and global provisioned throughput units. Standard regional restricts inferencing to one region; data zone (US, EU, or Asia-Pacific) allows any cluster within that geopolitical boundary for data sovereignty needs; global allows any cluster worldwide, often improving performance since inferencing latency typically outweighs network latency for non-streaming use. _Engineers architecting Azure AI deployments can dig into these throughput tradeoffs on daily.dev._ ## Similar posts on daily.dev - [Building AI apps and agents with Microsoft Foundry](https://daily.dev/posts/building-ai-apps-and-agents-with-microsoft-foundry-tdq7h47dl) · InfoWorld · 0 upvotes · 0 comments - [I Passed Microsoft’s AI-900. Then Microsoft Retired It](https://daily.dev/posts/i-passed-microsoft-s-ai-900-then-microsoft-retired-it-tt5vi8i0z) · Medium · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#azure](https://daily.dev/tags/azure) [View this post on daily.dev](https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"AI-103 Develop AI Apps and Agents on Azure Study Cram","url":"https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5"},"datePublished":"2026-08-17T10:46:02.755Z","dateModified":"2026-09-14T08:24:34.945Z","description":"A study guide for the Microsoft AI-103 certification (which replaces AI-102), walking through Azure's Microsoft Foundry platform for building generative AI...","image":"https://i.ytimg.com/vi/WK2BvjOYTCQ/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/WK2BvjOYTCQ/sddefault.jpg","isAccessibleForFree":true,"articleSection":"John Savill's Technical Training","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"John Savill's Technical Training","logo":"https://media.daily.dev/image/upload/s--4rApg0fT--/f_auto,q_auto/v1780213220/logos/john-savill?_a=BAMAMiWQ0","url":"https://daily.dev/sources/john-savill"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ai-agents,azure","timeRequired":"PT117M","video":{"@type":"VideoObject","name":"AI-103 Develop AI Apps and Agents on Azure Study Cram","description":"A study guide for the Microsoft AI-103 certification (which replaces AI-102), walking through Azure's Microsoft Foundry platform for building generative AI...","thumbnailUrl":"https://i.ytimg.com/vi/WK2BvjOYTCQ/sddefault.jpg","uploadDate":"2026-08-17T10:46:02.755Z","duration":"PT117M","url":"https://api.daily.dev/r/fX51Njoo5","embedUrl":"https://www.youtube.com/embed/WK2BvjOYTCQ"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"John Savill's Technical Training","item":"https://daily.dev/sources/john-savill"},{"@type":"ListItem","position":3,"name":"AI-103 Develop AI Apps and Agents on Azure Study Cram"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ai-103-develop-ai-apps-and-agents-on-azure-study-cram-fx51njoo5#faq","mainEntity":[{"@type":"Question","name":"What is the difference between the Foundry responses API and the OpenAI chat completions API in Azure AI Foundry?","acceptedAnswer":{"@type":"Answer","text":"The responses API is stateful, storing conversation history server-side so you only need to reference a previous response identifier for multi-turn interactions, and it works with both Foundry and OpenAI endpoints. Chat completions, by contrast, requires you to manually resend the full message array (including the system role message) on every call and only works with the OpenAI API, meaning context grows over time and you manage state yourself. Developers comparing Azure Foundry API options can track these workflow details on daily.dev."}},{"@type":"Question","name":"What is a model router in Azure AI Foundry and how does it choose which model to use?","acceptedAnswer":{"@type":"Answer","text":"A model router is a Foundry deployment that sits between an application and multiple underlying models (mini, nano, GPT variants, plus Llama, DeepSeek, Grok, and Anthropic when configured), using a small language model to automatically pick the best option for each request and route to it, optimizing cost without requiring custom logic. Similar behavior powers the auto option in Microsoft 365 Copilot and GitHub Copilot. Teams weighing multi-model strategies can follow model router developments on daily.dev."}},{"@type":"Question","name":"What are the deployment throughput options for models in Azure AI Foundry, and how do global versus data zone deployments differ?","acceptedAnswer":{"@type":"Answer","text":"Azure AI Foundry offers global standard, data zone standard, global batch, data zone batch, and global provisioned throughput units. Standard regional restricts inferencing to one region; data zone (US, EU, or Asia-Pacific) allows any cluster within that geopolitical boundary for data sovereignty needs; global allows any cluster worldwide, often improving performance since inferencing latency typically outweighs network latency for non-streaming use. Engineers architecting Azure AI deployments can dig into these throughput tradeoffs on daily.dev."}}]} ```