AI-103 case studies
AI-103 · Put your knowledge to work in context.
AI-103 leads to the Microsoft Certified: Azure AI Apps and Agents Developer Associate credential, an intermediate-level certification for AI engineers and developers who build, manage, and deploy agents and AI solutions that take advantage of Microsoft Foundry. Candidates should already be comfortable developing apps in Python and should understand the capabilities of general AI, generative AI, and Azure services. The role spans planning and managing Azure AI solutions plus implementing generative AI and agentic, computer vision, text analysis, and information extraction workloads, and it involves collaborating with business stakeholders, solution architects, data scientists, DevOps engineers, and cloud security engineers. The assessment is proctored, lasts 120 minutes, and may include interactive components alongside other item types. The AI-103 case studies are 5 cases and about 3,457 words.
1One resource or three: standing up a shared Azure AI platform
Contoso's platform team of two engineers must host three workloads in West Europe: a Document Intelligence pipeline at roughly 40,000 pages/month across 11 suppliers, an Azure OpenAI chat workload at a sustained 120,000 TPM, and Azure AI Vision tagging about 900,000 images/month. Security mandates private endpoints only, with no public egress from the spoke VNet.
707 words, about 4 min2HR assistant citing superseded policy and timing out
Fabrikam's HR assistant handles 6,000 questions/month over 42,000 documents. The leave policy exists in 14 revisions from 2019 onward, and all revisions are indexed.
713 words, about 4 min3Bottle-cap inspection with a thin dataset and no reliable network
Tailspin's bottling line rejects about 0.4% of bottles for cap misalignment. Three inspectors per shift currently do the check visually.
675 words, about 3 min4Support ticket routing with a 22% fallback rate
Northwind's support portal receives 85,000 tickets/month in English, German, and Japanese. The team built a Conversational Language Understanding (CLU) project with 9 intents and 12,000 utterances.
652 words, about 3 min5Supplier invoice extraction where line items do not sum to the total
Litware processes 14,000 supplier invoices/month across 14 suppliers. Layouts differ widely; three suppliers send faxed scans with handwritten totals.
710 words, about 4 min