Skip to main content

Posts

This week in AI updates: Mistral’s new Le Chat features, ChatGPT updates, and more (September 5, 2025)

Mistral announces new connectors, Memories Mistral announced that its generative AI chat Le Chat now connects with over 20 new connectors, including tools like Asana, Atlassian, Box, Databricks, GitHub, Outlook, Snowflake, Stripe, and Zapier. Users will also now be able to add their own connectors via MCP. The company also announced a beta for Memories, which allows users to set preferences to get more personalized responses. They can also import their memories from ChatGPT. Both of these features are available for any Le Chat user, including free users. OpenAI adds several minor updates to ChatGPT The company announced that users can now branch off conversations in ChatGPT to explore a specific direction while preserving the direction of the original thread. Additionally, Projects are now available to free users, and the company has added larger file uploads per project, the option to select colors and icons, and project-only memory controls. Google announces new open embeddin...
Post a Comment
Read more
Recent posts

Beyond the benchmarks: Understanding the coding personalities of different LLMs

Most reports comparing AI models are based on benchmarks of performance, but a recent research report from Sonar takes a different approach: grouping different models by their coding personalities and looking at the downsides of each when it comes to code quality. The researchers studied five different LLMs using the SonarQube Enterprise static analysis engine on over 4,000 Java assignments. The LLMs reviewed were Claude Sonnet 4, OpenCoder-8B, Llama 3.2 90B, GPT-4o, and Claude Sonnet 3.7. They found that the models had different traits, such as Claude Sonnet 4 being very verbose in its outputs, producing over 3x as many lines of code as OpenCoder-8B for the same problem. Based on these traits, the researchers divided the five models into coding archetypes. Claude Sonnet 4 was the “senior architect,” writing sophisticated, complex code, but introducing high-severity bugs. “Because of the level of technical difficulty attempted, there were more of these issues,” said Donald Fischer,...
Post a Comment
Read more

Neo4j introduces new graph architecture that allows operational and analytics workloads to be run together

The graph database company Neo4j today announced Infinigraph, a new distributed graph architecture that allows Neo4j’s database to run both operational and analytical workloads in one system. According to the company, silos often keep these workloads separate, leading AI applications to suffer, decision-making to be delayed, and costs to increase as a result of complex integration. Currently, some of the workarounds companies go through to bring these workloads together include having one database and one copy of data, one database and two engines (column-based and row-based), or having two or more synchronized databases. “Infinigraph eliminates the need for these workarounds. It enables organizations to run both analytical and transactional workloads in the same system, at unprecedented scale, while avoiding ETL pipelines, sync delays, and redundant infrastructure,” the company wrote in a blog post . Some examples of use cases that Infinigraph unlocks include the ability to detect ...
Post a Comment
Read more

Kong Acquires OpenMeter to Unlock AI and API Monetization for the Agentic Era

Kong Inc ., a leading developer of cloud API and AI technologies, today announced the acquisition of OpenMeter , a leading open-source and SaaS platform for usage-based metering and billing. The acquisition will bring usage-based monetization capabilities to Kong Konnect, the unified API platform. This will enable organizations to productize and bill for their APIs, AI, and data streams, seamlessly turning digital assets into new sources of revenue. In the AI era, billing becomes a metering problem. Agents will exchange labor via APIs. Large Language Models (LLM) – if not chats – are already sold via APIs. This requires a new kind of platform that brings together API infrastructure and monetization all in one. A unified approach to enforcement, security, and monetization. As AI adoption accelerates, digital connections are no longer deterministic or limited to the pace of human activity. Instead, they are continuous, machine-driven, and orders of magnitude larger. AI agents can trigg...
Post a Comment
Read more

Cloudsmith launches ML Model Registry to provide a single source of truth for AI models and datasets

Cloudsmith, providers of an artifact management platform, announced its ML Model Registry , which can act as a single source of truth for all AI models and datasets a company is using. The registry integrates with the Hugging Face Hub and SDK so that developers can push, pull, and manage models and datasets from Hugging Face and then use Cloudsmith to maintain centralized control, compliance, and visibility. Once data has been pushed from Hugging Face to Cloudsmith, security and compliance data can be utilized by Enterprise Policy Management so that teams can apply consistent policies to automatically quarantine, block, and approve specific models. It can also integrate with training, validation, and deployment pipelines, and provides protection of proprietary models and datasets via fine-grained access controls, entitlement tokens, and audit trails. Models and datasets are also managed in the same repositories as a company’s other artifacts, and can be organized by project, enviro...
Post a Comment
Read more

Microsoft Graph CLI to be retired

Microsoft has announced it is going to be retiring the Microsoft Graph CLI, with a deprecation phase starting now and full retirement scheduled for August 28th, 2026. During the deprecation phase, Microsoft will not add any new features and will only address critical vulnerabilities. According to Microsoft, this change is part of the company’s efforts to streamline the developer experience for Microsoft Graph by focusing its attention on PowerShell. The company recommends that users begin switching over to the Microsoft Graph PowerShell SDK, which offers broad API coverage and regular updates, integration with scripting and automation workflows, community and documentation support, and long-term support with Microsoft’s servicing commitments. The company explained that it initially released the CLI to offer a lightweight, cross-platform tool that developers could use to interact with the Microsoft Graph APIs. However, it was experiencing declining usage due to its limited extensibi...
Post a Comment
Read more

The state of DevOps and AI: Not just hype

Talk to any DevOps vendor today, and they’ll proudly tell you about their AI roadmap. Most vendors have already built something that will tick the checkbox, if that’s among your requirements. But checkboxes don’t solve problems. A feature that’s hard to use or adds extra manual steps to a developer’s processes doesn’t save you anything — and may end up costing you more than you expect. Just like you, vendors today are at the start of their AI journey. In some cases, the proof of concept gets packaged and shipped. The box is checked, the product goes out the door, and now it’s up to you to figure out if it’s worth using. Most DevOps AI Tools Are Still Point Solutions The truth is that nobody’s using one AI solution to address the entire software development lifecycle (SDLC). The vision of AI that takes you from a list of requirements through work items to build to test to, finally, deployment is still nothing more than a vision. In many cases, DevOps tool vendors use AI to build solu...
Post a Comment
Read more

UI automation: Why “try, try again”is your mantra

Over the weekend, I was looking to book flights for holidays. All excited, I was able to find a decent price for my family to head for some sunshine later this summer. I reached the payment stage, and that’s where the excitement and smile stopped. I was waiting for confirmation that my payment was finalizing, but then the system took a turn, and I was back on the booking page. Even more, I now had a higher price to pay for the flights. Did I continue with booking, NOOOOOO. The website had let me down, taking away all the joy and excitement I had built up. I just shut the laptop and cried in a corner.  (#WillDamienGoOnHolidays) UI automation promises efficiency, reliability, and a significant reduction in testing headaches. But let’s be honest, the road to seamless automation is not always smooth. If you have ever found your meticulously crafted test scripts failing unexpectedly, you’re not alone. In the world of UI automation, the phrase, “If at first you don’t succeed, try, try a...
Post a Comment
Read more

August 2025: All AI updates from the past month

Anthropic starts testing a Claude extension for Chrome The extension will enable Claude to take action on websites on behalf of the user. “We’ve spent recent months connecting Claude to your calendar, documents, and many other pieces of software. The next logical step is letting Claude work directly in your browser,” the company says. The company is starting off with a small pilot of 1,000 Max plan users, and will gradually expand the program out to more people if the pilot goes well. According to Anthropic, one of the big safety challenges with agents that use the browser is prompt injection attacks, and some of the steps the company has taken to defend against them are providing site-level permissions and requiring action confirmations. This pilot will test how well those defenses hold up in real-world scenarios. Google integrates Gemini CLI into Zed code editor Google announced that it has brought the Gemini CLI to the open source code editor, Zed. The new integration will en...
Post a Comment
Read more

Google integrates Gemini CLI into Zed code editor

Google has announced that it has brought the Gemini CLI to the open source code editor, Zed. The new integration will enable Zed users to generate and refactor code in the editor, get instant answers on code or error messages, and chat naturally in the terminal. Developers will be able to follow along live with the Gemini agent as it makes changes. Once the agent is done working, Zed will display the changes in a review interface that shows a clear diff for each edit that can be reviewed, accepted, or modified, providing the same level of control as a code review. Users will also be able to provide context beyond the codebase by pointing the agent to external sources like a URL with documentation or an API spec. This integration is possible because Zed created the Agent Client Protocol (ACP) to enable third-party agents to work directly in the platform. “Because software developers rely on diverse tools in a variety of different tech stacks, we see room for multiple agents comp...
Post a Comment
Read more

Microsoft packs Visual Studio August update with smarter AI features

Microsoft has released the August update for Visual Studio 2022, adding several features related to AI-assisted development. The company announced that GPT-5 is now integrated into the IDE, and support for MCP is generally available as well. MCP support enables developers to authenticate with any OAuth provider directly from the IDE, perform one-click installation of MCP servers, and manage MCP access from GitHub policy settings. Copilot Chat was updated with the ability to surface relevant code snippets more reliably using improved semantic code search to determine when queries should trigger a code lookup. Developers can now connect models from OpenAI, Google, and Anthropic to Visual Studio Chat, as well. Copilot Chat also now has full context of the user’s Git history, allowing developers to reference commits and uncommitted changes when engaging with the chat functionality. Another new update is that developers can sign up for GitHub Copilot using their Google account directly ...
Post a Comment
Read more

Amplitude launches new self-service capabilities for marketing initiatives

The product analytics company Amplitude is announcing new self-service capabilities that will enable marketing and growth teams to run user experiments without waiting on development teams. “Every click and conversion matters, and marketers can’t afford to wait weeks just to test an idea. Unfortunately, that’s the reality with most legacy experimentation tools: visual editors break on modern sites, simple changes require developer support, and even when a test runs, the results don’t connect back to the metrics you actually use,” Amplitude wrote in a blog post . The company is launching four new Web Experimentation features, including rearrange elements, control variant editing, group cohort targeting, and out-of-the-box widgets. Customers will now be able to visually move elements around on a page themselves using a drag-and-drop editing style. Control variant editing allows marketing teams to update the default user experience themselves. Marketers will also now be able to targe...
Post a Comment
Read more

Google locking down Android security with upcoming developer verification requirements for sideloaded apps

Android will soon require app developers to go through an identity verification process before their apps can be installed on users’ devices—regardless of if the apps are downloaded through the Play Store or sideloaded. “Think of it like an ID check at the airport, which confirms a traveler’s identity but is separate from the security screening of their bags; we will be confirming who the developer is, not reviewing the content of their app or where it came from,” Suzanne Frey, VP of Product, Trust & Growth for Android, wrote in a blog post . Google will be starting an early access program for verification in October 2025, where participants can join a community discussion forum, get priority support for the new rules, and be able to provide feedback. Then, starting in March 2026, all developers will be able to start verification. Verification requirements will go into effect in September 2026 in Brazil, Indonesia, Singapore, and Thailand, and then expand to other countries the...
Post a Comment
Read more

Microsoft donates DocumentDB to the Linux Foundation

Today at the Open Source Summit Europe, The Linux Foundation announced that the open-source document database, DocumentDB, would be joining the foundation and be released under the MIT license. DocumentDB was created by Microsoft and launched earlier this year. Since its release, it has gained 1.9k stars and hundreds of contributions, feedback, and users, according to the Linux Foundation. It started as a pair of PostgreSQL extensions to support popular BSON data models and document queries in the PostgreSQL ecosystem. With the recent addition of a gateway protocol transition layer, it has since evolved into a MongoDB compatible solution as well. Microsoft says that it is donating the project to the Linux Foundation for a couple of reasons. First, the foundation will help establish a standard for document databases, similar to what ANSI SQL did for relational databases. Second, it will provide a place for any database provider to continue. And finally, open-source Postgres will be f...
Post a Comment
Read more

This week in AI updates: Gemini Code Assist Agent Mode, GitHub’s Agents panel, and more (August 22, 2025)

Agent Mode in Gemini Code Assist now available in VS Code and IntelliJ This mode was introduced last month to the Insiders Channel for VS Code to expand the capabilities of Code Assist beyond prompts and responses to support actions like multiple file edits, full project context, and built-in tools and integration with ecosystem tools. Since being added to the Insiders Channel, several new features have been added , including the ability to edit code changes using Gemini’s Inline diff, user-friendly quota updates, real-time shell command output, and state preservation between IDE restarts. Separately, the company also announced new agentic capabilities in its AI Mode in Search, such as the ability to set dinner reservations based on factors like party size, date, time, location, and preferred type of food. U.S. users opted into the AI Mode experiment in Labs will also now see results that are more specific to their own preferences and interests. Google also announced that AI Mode i...
Post a Comment
Read more