Microsoft shared that user feedback on GitHub Copilot indicated that some developers struggle with creating prompts, including understanding phrasing and what context to include.  “In some cases, the experience left users feeling like they were getting too much or too little from their interactions,” Microsoft wrote in a blog post. 

In response to this, GitHub Copilot’s chat will now be a more conversational experience that can adapt to a developer’s specific context and needs. 

For example, if a developer asks a question that is too vague, like “what is this?,” Copilot will now respond back saying that the “question is ambiguous because it lacks specific context or content” and will suggest some prompts that are more specific and will lead to better responses. In this example, the response included other sample prompts, like “What is the purpose of the code in #file:’BasketService.cs’?” or “Can you explain the errors in #file:’BasketService.cs’?”

Those suggested new prompts are clickable, so all a developer has to do is select one of the provided prompts, and GitHub Copilot will try again with the new prompt.  

“Our guided chat experience takes Copilot beyond simple input-output exchanges, turning it into a collaborative assistant. When the context is clear, Copilot provides direct and relevant answers. When it isn’t, Copilot guides you by asking follow-up questions to ensure clarity and precision,” Microsoft wrote.

The new chat experience is available in Visual Studio 2022 17.12 Preview 3 and above, according to Microsoft.

The post GitHub Copilot chat now provides guidance on rewording prompts appeared first on SD Times.

By integrating GitHub Copilot capabilities into the tool, users will now be able to clean and transform data in VS Code with natural language prompts. It will also be able to provide suggestions of how to fix errors in data transformation code. 

According to Microsoft, one of the current limitations of using AI for exploratory data analysis is that the AI often lacks context of the data, leading to more generalized responses. Further, the process of verifying that the generated code is correct can be a very manual and time-consuming process. 

The integration of Data Wrangler and GitHub Copilot addresses these issues because it allows the user to provide GitHub Copilot with data context, enabling the tool to generate code for a specific dataset. It also provides a preview of the behavior of the code, which allows users to visually validate the response. 

Some examples of how GitHub Copilot can be used in Data Wrangler include formatting a datetime column, removing columns with over 40% missing values, or fixing an error in a data transformation — all using natural language prompts. 

Using this new integration will require having the Data Wrangler VS Code extension, the GitHub Copilot VS code extension, and an active GitHub Copilot subscription.

Microsoft also announced that this is just the first of many Copilot enhancements planned for Data Wrangler, and additional functionality will be added in the future. 

The post Microsoft enhances Data Wrangler with the ability to prepare data using natural language with new GitHub Copilot integration appeared first on SD Times.

The company has announced that the ability for users to create their own autonomous agents in Copilot Studio is moving from private to public preview next month.

Agents can be triggered by specific events and act on their own, rather than being activated by a conversation. For instance, when an email arrives, an agent can be activated to look up the sender’s details and account, see previous communications, check inventory, ask the sender their preferences, and then take the necessary actions to close a ticket. 

“Think of agents as the new apps for an AI-powered world. Every organization will have a constellation of agents — ranging from simple prompt-and-response to fully autonomous. They will work on behalf of an individual, team or function to execute and orchestrate business processes. Copilot is how you’ll interact with these agents, and they’ll do everything from accelerating lead generation and processing sales orders to automating your supply chain,” Microsoft wrote in its announcement. 

The company shared some examples of customers that have already built their own autonomous agents, including Pets at Home, who created an agent to compile cases for human review; McKinsey & Company, who created an agent to speed up client onboarding; and Thomson Reuters, who created an agent to speed up its legal due diligence process.

Microsoft is also releasing ten new autonomous agents in Dynamics 365, which is its enterprise resource planning (ERP) and customer relationship management (CRM) software.

New agents spans sales, service, finance, and supply chain use cases, and include examples like:

• Sales Qualification Agent, which researches leads, prioritizes opportunities, and guides customer outreach
• Supplier Communications Agent, which tracks supplier performance to detect and respond to delays
• Customer Intent and Customer Knowledge Management Agents, which learn from customer service representatives how to resolve customer issues and add articles to a company’s knowledge base. 

The post Microsoft 365 Copilot allows users to create their own autonomous agents appeared first on SD Times.

Passkeys are a safer alternative to passwords where users can authenticate using biometrics, a PIN, or pattern, and have been adopted by a number of other technology companies, like Amazon, GitHub, Google, PayPal, and more. 

With the new API support for third-party passkey providers, users will be able to utilize their preferred external passkey app. To implement this, Microsoft has been partnering with providers like 1Password, Bitwarden, and others.

“At Microsoft, we truly believe that security is a team sport. By partnering with OEMs, app developers and other partners in the ecosystem, and by helping people be better at protecting themselves—we are continuing to make Windows more secure by design and more secure by default,” Katharine Holdsworth, partner group product manager for Windows Security at Microsoft, wrote in a blog post. 

The experience for creating and using passkeys has also been improved. When a user visits a website that supports passkeys, they will be prompted about how they want to save those passkeys, with one of the options being to save to their Microsoft account. Users saving to their Microsoft account will need to go through a one-time setup where they will create a recovery key that will later be used to verify their identity and protect their passkeys. Once this initial setup is complete, they will be able to sign in to the website and save their passkey through Windows Hello. 

Passkeys saved through Windows Hello can be synced to multiple Windows 11 devices, providing users a more seamless sign-on process across all of their computers. 

“Microsoft is committed to making passkeys more readily available wherever you need them, with the experience, flexibility and durability that you expect when using Windows,” Holdsworth wrote. 

According to Microsoft, these three new features will start rolling out to the Windows Insider channels over the next several months. The company will also be sharing more information during sessions at Authenticate 2024 from October 14 to 16. 

The post Microsoft makes improvements to the passkey experience on Windows 11 appeared first on SD Times.

This new project “simplifies the automation of intelligent reactions in dynamic systems, delivering real-time actionable insights without the overhead of traditional data processing methods,” Mark Russinovich, CTO, deputy chief information security officer, and technical fellow at Microsoft Azure, wrote in a blog post. 

It watches for events in logs and change feeds without having to copy data to a central data lake or continuously querying data sources. Developers can use define which changes they want to track, and then Drasi decides if changes should trigger an action. 

“If they do, it executes context-aware reactions based on your business needs. This streamlined process reduces complexity, ensures timely action while the data is most relevant, and prevents important changes from slipping through the cracks,” Russinovich explained.

Drasi can be boiled down into three components: Sources, Continuous Queries, and Reactions. 

Sources connect to data sources like application logs, database updates, or system metrics and continuously monitor for critical changes.

Continuous Queries continuously evaluate incoming changes based on some predefined criteria.

Reactions are made when a change completes a continuous query, and can include tasks like sending alerts, updating other systems, or performing remediation steps. 

According to Microsoft, developers building event-handling mechanisms have often turned to multiple tools, resulting in fragmented and complex architectures. Additionally, a lot of change detection tools don’t have real-time capabilities and instead rely on batch processing, data collation, or delayed event analysis. 

“For businesses that need immediate reactions, even these slight delays can lead to missed opportunities or risks. In short, there is a pressing need for a comprehensive solution that detects and accurately interprets critical events, and automates appropriate, meaningful reactions,” Russinovich wrote.

The project has been submitted to the Cloud Native Computing Foundation (CNCF) as a Sandbox project, meaning that if it’s accepted, it will get support, guidance, and resources from the organization. It is being licensed under the Apache 2.0 license.

The post Microsoft open-sources Drasi, a data processing system for detecting and reacting to changes appeared first on SD Times.

SharePoint Framework enables developers to build applications or experiences in other Microsoft 365 apps that utilize data from SharePoint. 

The 1.20 preview includes two main additions, both related to building for the employee engagement platform Microsoft Viva. 

First, it is introducing new data visualization options — bar, pie, and donut chart types — for the Viva Connections card view. 

This could allow for dashboards that include a donut chart that shows departments and their budgets or a bar chart that shows the company’s sales distribution across different products, Sri Harsha Chilukuri, a software engineer at Microsoft, explained in a video presentation. 

Next, it added support for HTML Quick Views for Viva Connections cards, which, as the name implies, provide a quick view of information, such as a listing of recent files.

“The main idea here is that when you are on the go on your mobile, you can easily access the relevant information through the dashboard, and then as needed you can transition to the actual application for more advanced operations,” Vesa Juvonen, principal product manager at Microsoft, explained in a video.

According to Microsoft, this enables more flexible and complex layout options across desktop, tablet, and mobile devices.

“As with previous preview releases, our objective is to provide more insights into the upcoming features and to provide our worldwide ecosystem an option to directly influence on the released capabilities. Your feedback is welcome,” Juvonen wrote in a blog post. 

The post Microsoft shares first public preview of SharePoint Framework 1.20 appeared first on SD Times.

With this new experimental framework, agents are now supported in the platform. According to Microsoft, agents are AI entities that specialize in being able to complete a specific task with a specific set of tools. Multiple agents can work together to complete a complex task as well.

Within Semantic Kernel, each agent can be associated with different plug-ins and functions that can help it execute its job. 

The Agent Framework includes two specific agent types as well as abstractions that developers can build on. The first agent type is a Chat Completion Agent and the second is the Open AI Assistant Agent. 

There are also two modes for interacting with agents: directly invoking it or by chatting with it through AgentChat.

Upcoming Agent Framework features that are being worked on right now include support for the Open AI Assistant v2, enabling serialization and restoring AgentChat, improved chat patterns, and more. 

The Semantic Kernel project itself was first previewed in April 2023 and since then it has grown to 20K stars on GitHub. 

Read about other recent Open-Source Projects of the Week:

• • Hoppscotch
  • DevPod
  • GraphRAG

The post Semantic Kernel’s Agent Framework – SD Times Open Source Project of the Week appeared first on SD Times.

Now, the company is revealing that its Semantic Kernel team has been working on upgrading its connectors to use version 2 of the OpenAI library and Azure.AI.OpenAI library. 

According to the company, there were significant updates to the underlying APIs in the upgrade from v1 to v2, which is going to result in breaking changings that might impact Semantic Kernel developers using the library. 

Abstractions in Semantic Kernel isolate code from a majority of the changes, but there are still some that are unavoidable. Developers will need to update the name of the library they are importing because the names of the Semantic Kernel connectors have been updated to reflect that there are now two libraries that connect to OpenAI models. The new names are Microsoft.SemanticKernel.Connectors.OpenAI and Microsoft.SemanticKernel.Connectors.AzureOpenAI.

Other changes that may need to be made can be found in Microsoft’s blog post here. 

“Uptalking a major update can be challenging, but we in the Semantic Kernel team want to make it as painless as possible. As we get closer to adopting the new v2 libraries, we will provide a detailed migration guide to help you with the process of upgrading your code,” Mark Wallace, principal software engineer for Semantic Kernel, wrote in the blog post. 

You may also like…

The evolution and future of AI-driven testing: Ensuring quality and addressing bias

OpenAI taking on Google Search with prototype of SearchGPT

The post Microsoft provides guidance for upcoming support of OpenAI library v2 in Semantic Kernel appeared first on SD Times.

The software update only affected Windows operating systems; CrowdStrike instances running on Linux and Mac did not cause problems. 

Because the use of CrowdStrike and Windows is so prevalent among businesses, the outages were widespread, affecting several major airlines that had to delay/cancel flights, 911 operations, healthcare facilities, and more. 

“The current event appears – even in July – that it will be one of the most significant cyber issues of 2024. The damage to business processes at the global level is dramatic,” said Omer Grossman, CIO at CyberArk.

CrowdStrike CEO George Kurtz said in an X post that a fix for the issue had been made available. “This is not a security incident or cyberattack,” he wrote. “The issue has been identified, isolated and a fix has been deployed. We refer customers to the support portal for the latest updates and will continue to provide complete and continuous updates on our website. We further recommend organizations ensure they’re communicating with CrowdStrike representatives through official channels. Our team is fully mobilized to ensure the security and stability of CrowdStrike customers.”

Satya Nadella, CEO of Microsoft also said that it was working closely with CrowdStrike to help get customers back online.

Even though there is a fix available, it could still take days for these outages to resolve. “It turns out that because the endpoints have crashed – the Blue Screen of Death – they cannot be updated remotely and this problem must be solved manually, endpoint by endpoint,” said Grossman.

This event highlighted the problem with the majority of companies relying on just a few large technology vendors, such as Windows. According to Omkhar Arasaratnam, general manager of the Open Source Security Foundation (OpenSSF), these monocultural supply chains are inherently fragile. 

“Good system engineering tells us that changes in these systems should be rolled out gradually, observing the impact in small tranches vs. all at once,” said Arasaratnam. “More diverse ecosystems can tolerate rapid change as they’re resilient to systemic issues.”

Marcus Merrell, principal test strategist at Sauce Labs, agrees that an update like this should have been rolled out slowly over a period of several hours or days rather than “risk crippling the entire planet with one large update.”

He continued, “Everything is software and software is everything – it’s more interconnected and interdependent than ever. If the software update release going out there affects not just your users but your users ‘ users, you must  slow-roll the release over a period of hours or days, rather than risk crippling the entire planet with one large update.”

He also believes this outage highlights the need for better software quality. A recent survey from Sauce Labs found that 67% of respondents had at some point pushed code to production before testing it, and 28% say they do that regularly. 

According to Merrell, companies need to assess the risks vs benefit of any potential release. “The equation is simple: what is the risk of not shipping a code versus the risk of shutting down the world,” he said. “The vulnerabilities fixed in this update were pretty minor by comparison to ‘planes don’t work anymore’, and will likely have the knock-on effect of people not trusting auto-updates or security firms full stop, at least for a while.”

You may also like…

The secret to better products? Let engineers drive vision

Microsoft gives up its observer seat on OpenAI’s board

The post Bad CrowdStrike update takes down Windows machines around the world, highlighting importance of gradual roll-outs and software quality appeared first on SD Times.

RAG is an approach in which data is fed into an LLM to give more accurate responses. For instance, a company might use RAG to be able to use its own private data in a generative AI app so that employees can get responses specific to their company’s own data, such as HR policies, sales data, etc. 

How GraphRAG works is that the LLM creates the knowledge graph by processing the private dataset and creating references to entities and relationships in the source data. Then the knowledge graph is used to create a bottom-up clustering where data is organized into semantic clusters. At query time, both the knowledge graph and the clusters are provided to the LLM context window. 

According to Microsoft researchers, it performs well in two areas that baseline RAG typically struggles with: connecting the dots between information and summarizing large data collections. 

As a test of GraphRAG’s effectiveness, the researchers used the Violent Incident Information from News Articles (VIINA) dataset, which compiles information from news reports on the war in Ukraine. This was chosen because of its complexity, presence of differing opinions and partial information, and its recency, meaning it wouldn’t be included in the LLM’s training dataset. 

Both the baseline RAG and GraphRAG were able to answer the question “What is Novorossiya?” Only GraphRAG was able to answer the follow-up question “What has Novorossiya done?”

“Baseline RAG fails to answer this question. Looking at the source documents inserted into the context window, none of the text segments discuss Novorossiya, resulting in this failure. In comparison, the GraphRAG approach discovered an entity in the query, Novorossiya. This allows the LLM to ground itself in the graph and results in a superior answer that contains provenance through links to the original supporting text,” the researchers wrote in a blog post.  

The second area that GraphRAG succeeds at is summarizing large datasets. Using the same VIINA dataset, the researchers ask the question “What are the top 5 themes in the data?” Baseline RAG returns back five items about Russia in general with no relation to the conflict, while GraphRAG returns much more detailed answers that more closely reflect the themes of the dataset. 

“By combining LLM-generated knowledge graphs and graph machine learning, GraphRAG enables us to answer important classes of questions that we cannot attempt with baseline RAG alone. We have seen promising results after applying this technology to a variety of scenarios, including social media, news articles, workplace productivity, and chemistry. Looking forward, we plan to work closely with customers on a variety of new domains as we continue to apply this technology while working on metrics and robust evaluation. We look forward to sharing more as our research continues,” the researchers wrote.

Read about other recent Open-Source Projects of the Week:

• Theia IDE
• LibreChat
• Unity Catalog 

The post GraphRAG – SD Times Open Source Project of the Week appeared first on SD Times.