Hello,

I am currently in the process of onboarding a set of Windows and Linux servers into Defender for Servers. 

I am trying to figure out how I can pull the Device Inventory list located under Microsoft Defender -> Assets -> Device Inventory programmatically.

I tried Get-MgDeviceManagementManagedDevice but this only returns the devices in Intune which does not include the servers.

Thanks

​Hello,I am currently in the process of onboarding a set of Windows and Linux servers into Defender for Servers. I am trying to figure out how I can pull the Device Inventory list located under Microsoft Defender -> Assets -> Device Inventory programmatically.I tried Get-MgDeviceManagementManagedDevice but this only returns the devices in Intune which does not include the servers.Thanks   Read More

​

Hi all.  I am new and apologize if this has been asked (I looked but could not find this issue exactly).  I just switched to New Outlook and it appears that I can’t manually rearrange subfolders within a parent folder.  When I click-drag, I only see the visual feedback for moving it INTO another folder.  In Old Outlook, I could see visual feedback of a line BETWEEN other folders as well, and I could drop it on the line to effectively change the sort order manually.  Am I missing something?

​Hi all.  I am new and apologize if this has been asked (I looked but could not find this issue exactly).  I just switched to New Outlook and it appears that I can’t manually rearrange subfolders within a parent folder.  When I click-drag, I only see the visual feedback for moving it INTO another folder.  In Old Outlook, I could see visual feedback of a line BETWEEN other folders as well, and I could drop it on the line to effectively change the sort order manually.  Am I missing something?  Read More

​

We are finding that our scheduled news posts in SharePoint online are appearing in Viva Connections before the publish date/time in SharePoint. Is there a way to prevent this early posting?

​We are finding that our scheduled news posts in SharePoint online are appearing in Viva Connections before the publish date/time in SharePoint. Is there a way to prevent this early posting?  Read More

​

As we navigate the rapidly evolving landscape of Artificial Intelligence (AI), one skill has emerged as a critical component in harnessing the true potential of AI systems: Prompt Engineering. Generative AI, powered by Large Language Models (LLMs) like OpenAI’s GPT (Generative Pre-trained Transformer) series, has revolutionized the way we create content—be it text, images, audio, or even code. At the heart of this transformation lies the art of crafting and optimizing prompts to elicit consistent and high-quality responses from these models.

What is Prompt Engineering?

 Prompt Engineering is the process of designing and refining text inputs (prompts) to achieve specific application objectives with AI models. Think of it as a two-step journey:

Designing the Initial Prompt: Creating the initial input for the model to achieve the desired result.
Refining the Prompt: Continuously adjusting the prompt to enhance the quality of the responses.

Why is Prompt Engineering Important?

To understand the significance of Prompt Engineering, we need to delve into three key concepts:

Tokenization: The method by which the model perceives the prompt.
Base LLMs: The process through which the foundational model handles the prompt.
Instruction-Tuned LLMs: The way the model understands and executes tasks.

Key Reasons for the Importance of Prompt Engineering

Stochastic Nature of Model Responses: AI responses are inherently variable. The same prompt can produce different outcomes depending on the model or even at different times with the same model. Effective prompt engineering can reduce these variations by establishing more precise guidelines.
Mitigating Fabrications: Since models are trained on large yet finite datasets, they can sometimes generate inaccurate or fictitious responses. Prompt engineering techniques can help detect and address these fabrications by, for instance, requesting citations or reasoning from the AI.
Adapting to Model Capabilities: Newer models offer enhanced capabilities but also come with unique quirks and trade-offs in terms of cost and complexity. Prompt engineering helps create best practices that accommodate these model-specific characteristics, ensuring a scalable and efficient workflow.

Elements of a Prompt

A well-crafted prompt comprises several elements:

Context: Background information that sets the stage for the task.
Instructions: Clear directives guiding the AI on what to do.
Exemplary Input: Examples that illustrate the task.
Primary Content: The main input data for which a response is sought.
Output Indicator: Signals indicating the desired output format.
Each element plays a vital role in guiding the AI system to perform specific tasks, such as providing a sentiment analysis for an input sentence.

 Techniques in Prompt Engineering

 Prompt Engineering employs a variety of techniques, each suited to different scenarios:

Zero-Shot Learning: The model generates responses without specific examples, relying solely on pre-training data. This is useful when labeled data is impractical or unavailable.
Few-Shot Learning: Providing the model with a few examples to help it understand and perform a specific task. This is beneficial when limited data is available or quick adaptation to new tasks is needed.
System Messages and Roles: Pre-set instructions that guide the AI’s responses, ensuring consistency and alignment with user goals. Assigning specific roles to the AI allows it to simulate different personas and provide contextually relevant responses.

Conclusion 
Prompt Engineering is a rapidly evolving field within AI that offers innovative ways to interact with AI systems. It represents a significant advancement in our ability to harness the power of AI, providing us with tools to achieve more accurate, reliable, and contextually appropriate responses. As we continue to explore and refine the techniques of Prompt Engineering, we are not only enhancing the capabilities of AI systems but also paving the way for more seamless human-AI collaboration.
The importance of Prompt Engineering cannot be overstated. It helps mitigate the inherent variability and potential inaccuracies in AI responses, ensures adaptability across different model capabilities, and unlocks new applications and use cases. By mastering this skill, we move closer to realizing the full potential of AI, making it an indispensable part of our technological future.
So, whether you’re an AI researcher, a developer, or simply an enthusiast, delving into the intricacies of Prompt Engineering is a worthwhile endeavor. It’s a testament to the incredible progress we’ve made in AI research and a glimpse into the exciting future that lies ahead. As we continue to refine our understanding and techniques, the possibilities are boundless, and the journey is just beginning.

​Microsoft Tech Community – Latest Blogs –Read More 

I saw this on LinkedIn. Is it true and where can we get more info? Gracias 

SQL Trace and SQL Server Profiler are deprecated. The Trace namespace that contains the Microsoft SQL Server Trace and Replay objects are also deprecated.”

We’ve all seen the warning above. Use Extended Events instead.

Profiler and SQL Server Trace are hard tools to replace. Especially when you’ve been using them for years. And learned how to use them well enough to not bring down a production server in the middle of the day.

​I saw this on LinkedIn. Is it true and where can we get more info? Gracias    SQL Trace and SQL Server Profiler are deprecated. The Trace namespace that contains the Microsoft SQL Server Trace and Replay objects are also deprecated.”We’ve all seen the warning above. Use Extended Events instead.Profiler and SQL Server Trace are hard tools to replace. Especially when you’ve been using them for years. And learned how to use them well enough to not bring down a production server in the middle of the day.  Read More

​

Payment platform and Microsoft services bolster retail in the Middle East

Azure services enable sentiment analysis, monitoring insights, and more

Based in Dubai, United Arab Emirates, Spotii empowers the GCC retail community with its focus on technology, trust, and inclusion. Spotii joined the Microsoft ecosystem through the Microsoft for Startups program in February, and since going live to the public in April, more than 100 merchants, ranging from small and medium-sized businesses to enterprises, have committed to Spotii’s service. Spotii’s plugin, available on Microsoft AppSource, easily integrates into Magento, WooCommerce, Shopify, and other e-commerce platforms.

Additionally, Spotii’s development team can support custom integrations as well as in-store functionality.

Spotii uses Azure Kubernetes Service to host its web and data apps. Its web app depends on Azure Database for PostgreSQL to store its data. Azure Monitor Log Analytics allows Spotii to monitor its apps. Helpful insights, such as tracing error messages in the log, are provided by Azure Kubernetes Service. Spotii uses the optical character recognition capabilities of Azure Cognitive Services to extract user information from a submitted ID image. Text analytics and sentiment analysis are Azure Cognitive Services tools that Spotii uses to evaluate customer feedback when scraping Instagram, LinkedIn, Facebook, or other social media sites.

Continue reading here

**Explore all case studies or submit your own**

​Microsoft Tech Community – Latest Blogs –Read More