From Me
Welcome everyone!
In this issue, we will cover some of the exciting announcements and updates from the Microsoft Build 2024 conference, which took place from May 21 to May 23. Microsoft Build is the annual developer event where Microsoft showcases its latest innovations and technologies for developers and customers.
Here are some of the topics we will explore in this issue:
How Azure Open AI is enabling new scenarios and capabilities for GitHub Copilot, such as natural language queries, code refactoring, and code generation from sketches.
How Azure Open AI is powering new tools and services for developers, such as Azure Code, Azure AI Studio, and Azure Bot Framework.
How Azure Open AI is advancing the state of the art in natural language processing, computer vision, and reinforcement learning, with new models, datasets, and benchmarks.
How Azure Open AI is fostering a vibrant and diverse community of developers and researchers, with new programs, events, and resources.
Essentially, you could call this the #MSBuild issue, but make sure not to miss out on the non-Build content. I hope you enjoy this issue and find it useful and informative.
From My AI
This week I asked my AI to give me a brief history of the Microsoft Build conference. Here’s the results…
Microsoft Build is an annual conference event organized by Microsoft Corporation. It is primarily aimed at software engineers, developers, and other technology enthusiasts who are interested in the latest tools, technologies, and trends in software development.
The first Microsoft Build conference was held in 2011 in Anaheim, California. It was initially called the "Windows Developer Conference" and was focused primarily on developing applications for Windows 8. However, it was later renamed to Microsoft Build and has since become a major event for developers across various platforms.
Over the years, Microsoft Build has been the stage for a number of significant announcements and keynotes, including the unveiling of Windows 10, the Microsoft HoloLens, the Microsoft Bot Framework, and the Microsoft Azure IoT Suite. Developers can attend various sessions, keynotes, and workshops to learn more about the latest tools and technologies, and to network with other developers and industry experts.
Overall, the Microsoft Build conference has become an important event for the technology industry, and it continues to evolve and grow each year.
From Microsoft
Announcements from Microsoft Build 2024:
Azure AI Vision at Microsoft Build 2024: Multimodal AI for Everyone
New generative AI app evaluation and monitoring capabilities in Azure AI Studio
Announcing Fine Tuning for GPT-4, AI Studio Support, and API improvements with Azure OpenAI Service
Announcing Multimodal Innovations in Generative AI with Azure OpenAI Service: Microsoft Build 2024
Accelerate enterprise GenAI application development with tracing and debugging through prompt flow
Microsoft Build 2024: Create custom copilots from SharePoint
Expanding Horizons: Azure AI Models as a Service Now Integrated with Dataloop, Arize AI, and ClearML
Introducing the new Azure AI infrastructure VM series ND MI300X v5
Introducing GitHub Copilot for Azure: Your Cloud Coding Companion in VS Code!
Getting Started - Generative AI with Phi-3-mini: Running Phi-3-mini in Intel AI PC
Azure AI Search’s new hybrid and vector search updates to boost GenAI app performance
Control Model Catalog deployments with Azure RBAC and Azure Policy
Multilingual Chatbot with Azure AI Studio, Phi-3 Mini, GPT-4 and Azure AI Translator
Introducing GenAI Gateway Capabilities in Azure API Management
Why Should Business Adopt RAG and migrate from LLMs? - In this blog we are going to discuss the importance of migrating your product or startup project from LLMS to RAG. Adopting RAG empowers businesses to leverage external knowledge, enhance accuracy, and create more robust AI applications. It’s a strategic move toward building intelligent systems that bridge the gap between generative capabilities and authoritative information.
Preparing your organization for AI: Insights from Microsoft’s roll-out of Copilot in Viva Glint - On May 16, the Viva People Science team held the third webinar in its AI Empowerment series. During this webinar, I was joined by Carolyn Kalafut (Principal People Scientist at Microsoft Viva), and Ketaki Sodhi (Senior Analyst and Program Lead, Employee Listening at Microsoft).
Observability, security, and AI revolutionize DevSecOps and platform engineering on Microsoft Azure - In the fast-evolving landscape of cloud-native technology and software delivery, DevOps and platform engineering are foundational. Yet, to ensure reliability, security, and efficiency, these practices require additional support. One crucial investment is observability.
Azure OpenAI chat baseline architecture in an Azure landing zone - The excitement surrounding Microsoft Build 2024 brings into focus the significant steps required to integrate AI and ML innovations into Azure workloads effectively. An essential resource for this integration journey is the Azure OpenAI Chat Baseline Architecture within an Azure landing zone – a strategic framework crucial for transitioning generative AI from the experimental phase into production.
How To Control Azure OpenAI Models - The first step is to choose the correct model and configure its parameters appropriately. Depending on your use case, you need to determine which model can deliver the best quality. This requires knowledge of the available models and their capabilities. Once you have selected the right model, focus on the parameters. For instance, setting the temperature to 0 will prevent the model from generating creative responses. Another important parameter is the max_token parameter which is crucial for optimizing latency performance. For retrieval-augmented generation use cases, it can be set a value between 500 to 800 tokens as a best practice. This parameter also affects the number of API calls you can make per minute, so finding the optimal value is essential. Additionally, the stop_sequence parameter allows you to define when the model should stop. You need to review and adjust other parameters as needed to ensure they meet the requirements of your use case.
Evaluating Large and Small Language Models on Custom Data Using Azure Prompt Flow - In recent years, the field of Artificial Intelligence (AI) has witnessed remarkable advancements, leading to the unprecedented surge in the development of small and large language models. They’re at the heart of various applications, aiding in everything from customer service chatbots to content creation and software development. These models offer developers a plethora of options, catering to a wide array of applications. However, this abundance also introduces complexity in choosing the right model that not only delivers optimal performance on specific datasets but also aligns with business objectives such as cost efficiency, low latency, and content constraints.
Exploring the New Frontier of AI: OpenAI's GPT-4-o For Indic Languages - In the ever-evolving landscape of artificial intelligence, OpenAI has once again pushed the boundaries with the introduction of the GPT-4-o model, featuring the innovative o200k_base tokenizer. This development marks a significant leap forward in the field, offering unprecedented speed, affordability, and multimodal capabilities.
Optimizing Retrieval for RAG Apps: Vector Search and Hybrid Techniques - In this blog we are going to dive into optimizing our search strategy with Hybrid search techniques.
From Others
Roadmap to Learn AI in 2024 - To escape tutorial hell and really learn, you have to get hands-on, write algorithms from scratch, implement papers, and do fun side projects using AI to solve problems. This article attempts to craft a free curriculum that follows that philosophy. I’m working on some of these courses.
Revolutionizing Education: The Impact of Generative AI on Learning - Generative AI, a subfield of artificial intelligence, has the remarkable ability to create original content, from text and images to audio and even video. This remarkable capability has profound implications for education, as it enables the development of personalized learning experiences, adaptive teaching methods, and innovative pedagogical tools.
The AI Attack Surface Map v1.0 - A visual map for understanding attacks against AI and AI systems.
Use the Azure OpenAI GPT-4o (all-in-one-model) with PowerShell! - I have created a PowerShell script that keeps the context of the conversation, analyze images and get responses based on them. I have written the script so you can both use local images and URL from online images.
Copilot for Security stuff now has its own bi-weekly newsletter!
From the Industry
Microsoft unveils Copilot+ PCs with generative AI capabilities baked in - Microsoft is calling its version Copilot+ PCs, which CEO Satya Nadella described as a "new class of Windows PCs." These contain hardware designed to handle more generative AI Copilot processes locally, rather than relying on the cloud. Doing so requires a chipset with a neural processing unit (NPU), and manufacturers such as Qualcomm have been laying the groundwork with chips like the Snapdragon X Elite.
I Tried Microsoft’s New AI-Focused PCs. Windows Is Exciting Again. - Microsoft CEO Satya Nadella told our columnist new efficient, powerful chips equip Windows Copilot+ laptops to compete with Apple’s MacBooks.
Recall is Microsoft’s key to unlocking the future of PCs - Microsoft’s groundbreaking new AI-powered tool can search and recall anything you’ve seen or done on your PC like a personal historian.
Microsoft Paint is getting an AI-powered image generator that responds to your text prompts and doodles - Microsoft Paint is getting new image generation powers with a new tool called Cocreator. Powered by "diffusion-based algorithms," Cocreator can generate images based on text prompts as well as your own doodles in the Paint app.
Microsoft, OpenAI, Google and other major companies reach AI safety agreement - Sixteen companies acknowledged both the huge potential benefits of AI and the risks if it is misused or harms people. Practical steps agreed to include publishing safety frameworks, avoiding models with unmanageable risks, and coordinating with regulators internationally. The goal is to ensure AI systems behave ethically and avoid any unintentional harm as the technology evolves.
Khan Academy and Microsoft are teaming up to give teachers a free AI assistant - AI assistant Khanmigo can help time-strapped teachers come up with lesson ideas and test questions, the companies say.
Microsoft’s new Copilot AI agents act like virtual employees to automate tasks - Microsoft argues that its AI automation will remove the boring bits of jobs instead of replacing jobs entirely.