‹ All AI-103 questions

AI-103 Practice Question 96: Draft Version Promotion

Microsoft Azure AI Apps and Agents Developer Associate (AI-103) · Test 10 · question 6 of 10

Implement generative AI and agentic solutionsmedium

A team runs a Hosted agent in production; version 3 currently serves 100% of the endpoint's traffic. They want to test version 4, a new container image with an updated tool, through real interactive testing before any production traffic reaches it, without disrupting version 3's traffic and without advancing the release version number that client code or documentation might already reference.

How should they proceed?

Select one answer
Show answer and explanation

Answer:

  • B. Create the new image as a draft version, test it directly by its assigned draft-{timestamp} identifier, then create it again with the same configuration but without the draft flag once satisfied, so it becomes a normal release version

A draft version is built for exactly this: "an experimental version that you can create and test without affecting how your agent serves production traffic." It's assigned a draft-{timestamp} identifier instead of an auto-incremented number, so "it never advances your release version numbering," and it's "excluded from implicit routing" and "can't be a traffic-routing target," so it can't leak into production by accident. Once satisfied, you "promote it by creating a new version with the same image and configuration but with draft omitted... The new version receives a regular auto-incremented version number and becomes eligible for traffic routing."

- A is wrong because traffic splitting between agent versions isn't supported; an endpoint's version selector must configure one FixedRatio rule at 100%, so a 50/50 split isn't a configuration the platform allows. - C is wrong because it works but adds unforced overhead: a new agent means a new identity, a new endpoint, and a client change, all to get isolation the built-in draft mechanism already provides without touching version 3. - D is wrong because it takes version 3 fully offline for the duration of testing, exactly the production disruption the team is trying to avoid.

Reference: Manage hosted agents

Practice Test 10 ›

Study guides for AI-103

AI engineer roadmap 2026: skills, certs and the orderAI engineer roadmap for 2026: skills to learn in order, AI engineer vs ML engineer, which certifications fit each stage, and what to skip.
AI-200 exam guide: Azure AI Cloud Developer AssociateAI-200 replaced AZ-204: what the Azure AI Cloud Developer exam tests, how it differs from AI-103, price in USD and INR, format and how to prepare.
How AI Is Changing Software Jobs: Where to PrepareHow AI is changing software jobs: what history shows, the evidence on junior developers, what Klarna learned, new AI roles and five domains to prepare for.
AI-901 exam guide: Azure AI Fundamentals after AI-900AI-901 replaced AI-900: learn what changed, how much Python and Microsoft Foundry you need, exam costs in India, and how to prepare.
Best AI certifications 2026: AWS, Google, Microsoft, ClaudeThe best AI certifications in 2026 from AWS, Google, Microsoft, Claude and Databricks, by level, with price, status and which one to take first.
Microsoft AI certification path 2026: which exam firstMicrosoft AI certification path for 2026 - AI-901, AB-730, AI-103, AI-200, AI-300, AB-100, AI-500 and GH-300 by role, and which exam to take first.
AI-103 Certification Guide: What Changed From AI-102The AI-103 certification replaced AI-102 in 2026. Exam format, cost, the five skill domains, what changed and how to prepare, with free practice questions.
What Is RAG? How Retrieval-Augmented Generation WorksWhat RAG is and how it works: chunking, embeddings, vector search, hybrid search, reranking, GraphRAG and agentic RAG on AWS, Google and Azure.
Prompt engineering vs RAG vs fine-tuningWhen to use prompt engineering, retrieval-augmented generation, fine-tuning or continued pre-training, compared on cost, freshness, hallucination and privacy.