Implement natural language processing solutions (30–35%)
Implement knowledge mining and document intelligence solutions (10–15%)
Implement generative AI solutions (10–15%)
General Concepts
Responsible AI
Fairness: Ensure AI models treat all groups of people fairly.
Inclusiveness: Design AI systems that benefit everyone and respect human rights.
Reliability & Safety: Develop AI that is safe, reliable, and resilient.
Privacy & Security: Safeguard personal data and maintain user privacy.
Transparency: Make AI decisions understandable and transparent.
Accountability: Ensure that AI systems are accountable for their actions and decisions.
Selecting the Right Azure AI Services
Azure Computer Vision: Helps analyze images, detect objects, and extract text (OCR). I can use this for detecting image contents, generating image tags, and converting handwritten text into digital data. Good for visual content analysis.
Azure Language: Excellent for natural language processing (NLP) tasks, such as sentiment analysis, key phrase extraction, and entity recognition. I should focus on this when working with textual data that requires interpretation.
Azure Speech Services: Converts speech to text, text to speech, and supports speech translation in real-time. This will be useful in voice applications and AI assistants.
Azure OpenAI: Use this for generating human-like text or images from prompts. It’s great for natural language generation, summarization, or content creation.
Azure Document Intelligence: Good for extracting structured information from forms, invoices, and documents. It can handle both standard and custom forms, streamlining the data extraction process.
Azure Cognitive Services
Vision Services
Computer Vision
Image Analysis: Analyzes images to extract information such as objects, faces, and text. It supports operations like smart cropping, generating image descriptions, and tagging images.
Description: Detects and recognizes human faces in images and videos. It can identify facial attributes like age, emotion, and head pose, as well as perform face matching and verification.
Description: Extracts text, key/value pairs, and tables from documents, such as receipts, invoices, and business cards. It allows you to build custom models to extract specific information.
Description: Enables you to build custom image classifiers and object detectors. You can upload images, tag them, and train models to recognize specific objects or scenes.
Description: Extracts metadata and insights from video content, including face identification, text recognition, object detection, and scene segmentation.
Description: Helps build natural language understanding into apps, bots, and IoT devices by allowing them to understand user intents and context. (LUIS will be retired on October 1st 2025 and starting April 1st 2023 you will not be able to create new LUIS resources. We recommend migrating your LUIS applications to conversational language understanding to benefit from continued product support and multilingual capabilities.)
Description: Provides real-time text translation across multiple languages, supporting more than 60 languages for both standard and neural machine translation.
Description: Detects potentially offensive, risky, or unwanted content in text, images, and videos, providing automated moderation tools. Azure Content Moderator is deprecated as of February 2024 and will be retired by February 2027. It is replaced by Azure AI Content Safety, which offers advanced AI features and enhanced performance.
Description: Azure AI Search (formerly known as “Azure Cognitive Search”) provides secure information retrieval at scale over user-owned content in traditional and generative AI search applications. Information retrieval is foundational to any app that surfaces text and vectors. Common scenarios include catalog or document search, data exploration, and increasingly chat-style apps over proprietary grounding data.
Description: Creates a question-and-answer layer over your data, enabling natural language processing to respond to user queries from existing content like FAQs, manuals, and documents. The QnA Maker service is being retired on the 31st of March, 2025. A newer version of the question and answering capability is now available as part of Azure AI Language.
Description: Integrates powerful language models from OpenAI with Azure’s robust infrastructure, allowing you to implement advanced language understanding and generation capabilities. Azure OpenAI Service provides REST API access to OpenAI’s powerful language models including GPT-4o, GPT-4 Turbo with Vision, GPT-4, GPT-3.5-Turbo, and Embeddings model series.
Image Analysis: Using AI to analyze and understand the content of images.
Object Detection: Identifying and locating objects within images.
Text Analysis: Extracting meaningful insights from text data.
Language Translation: Translating text or speech from one language to another.
Speech Recognition: Converting spoken words into text.
Anomaly Detection: Identifying unusual patterns or behaviors in data.
Form Processing: Extracting data from forms and structured documents.
Bot Services: Creating conversational agents to interact with users.
Important Concepts to Keep in Mind
Responsible AI: Ensure fairness, transparency, and accountability in AI models. I’ll have to focus on understanding how to align AI solutions with Microsoft’s Responsible AI principles.
Monitoring & Cost Management: Set up proper monitoring and diagnostic tools for Azure AI services. Monitoring includes tracking performance and diagnosing issues, while cost management ensures the AI solutions stay within budget.