Tailwind Traders wants to embed generative AI into a custom line-of-business web application. Developers need API-level access to large language…

Azure OpenAI Service is the appropriate choice as it provides API-level access to foundational large language models (LLMs). This enables developers to build custom applications and maintain full control over prompt engineering. Crucially, it includes the "on your data" feature, a retrieval-augmented generation (RAG) mechanism that allows the model's responses to be grounded in the company's proprietary data sources. This directly meets all the specified requirements for building a custom, data-aware generative AI application.

A. Microsoft Copilot for Sales is a pre-built SaaS application for CRM integration, not a platform for building custom web apps from scratch.

B. Power Automate process mining is a tool for analyzing and optimizing business workflows, not for developing generative AI applications.

D. Microsoft Copilot for Microsoft 365 is an integrated assistant within Office apps; it is not the foundational service for building new, standalone web applications.

1. Microsoft Documentation, "What is Azure OpenAI Service?". Azure AI Services documentation. Retrieved October 10, 2023. States that the service "provides REST API access to OpenAI's powerful language models."

2. Microsoft Documentation, "Azure OpenAI on your data". Azure AI Services documentation. Retrieved October 10, 2023. Describes the feature that enables developers to "run supported chat models on your data" using retrieval-augmented generation.

3. Microsoft Learn, "Plan your AI transformation journey with Microsoft". AI Transformation Leader Learning Path. Module 2, Unit 3: "Build custom AI solutions". This module positions Azure AI services, including Azure OpenAI, as the platform for building bespoke AI solutions.