Articles by Category: Product_announcements

Evaluate your AI agents faster and more effectively

2025-12-04 16:20
🚀 Exciting updates for AI developers! The DigitalOcean Gradient™ AI Platform has enhanced its agent evaluations experience for easier and faster assessments. Key changes include: - **Goal-oriented metric grouping** for better organization of Safety & Security, Correctness, and RAG Performance. - **Example datasets** for quick dataset creation. - **Clear error messaging** to simplify debugging. - **Interpretable results** with trace integration for efficient performance analysis. Evaluations...
Grace Morgan

OpenJDK 25 now available in Red Hat Enterprise Linux 10.1

2025-12-04 16:07
🚀 Red Hat Enterprise Linux (RHEL) 10.1 now features OpenJDK 25 LTS, enhancing startup performance, memory efficiency, and developer ergonomics. 🔧 This update supports developers until December 2030, offering a robust foundation for applications. Key improvements include better garbage collection and concurrency features. 📦 To install, use: `sudo yum install java-25-openjdk` #OpenJDK #RHEL #JavaDevelopment #RedHat #SoftwareUpdates
Kangcheng Xu

Faster Agents with Automatic Prompt Caching

2025-12-04 16:00
🚀 Exciting news from Heroku! Starting December 18, 2025, they will introduce automatic prompt caching for Managed Inference and Agents. This feature, enabled by default, aims to enhance response times for common workloads without any extra effort. Currently, caching focuses on system prompts and tool definitions, excluding user messages. You can disable caching for specific requests if needed. Prompt caching works by securely storing and reusing processed components, leading to faster...
Source: Heroku Blog
Anush DSouza

Introducing OfficeQA: A Benchmark for End-to-End Grounded Reasoning

2025-12-04 15:30
🌟 Exciting news in AI research! The article introduces OfficeQA, a new benchmark designed for end-to-end grounded reasoning. This benchmark aims to enhance the evaluation of AI systems' capabilities in understanding and processing information. OfficeQA is expected to support advancements in natural language processing and reasoning tasks. #AI #MachineLearning #NaturalLanguageProcessing #Research #OfficeQA

Building Trust at Scale

2025-12-04 14:59
🔍 Palantir introduces “audit.3,” a next-generation audit logging system enhancing transparency and accountability across its platforms. This evolution aims to provide quicker, structured logs, vital for organizations managing sensitive data. Comprehensive audit logging helps security teams investigate incidents and ensures compliance. 📊 The new system improves log accessibility and consistency, allowing for real-time data analysis, boosting trust in the handling of sensitive information....
Palantir

Mooncake Brings Databricks Rich Transactional Processing

2025-12-04 14:00
Databricks recently acquired Mooncake, enhancing its Lakebase platform for AI agents. 🌐 Mooncake offers rich transactional processing and fast analysis without managing ETL pipelines. This integration allows direct data access from PostgreSQL for decision-making. Developed by ex-SingleStore engineers, Mooncake aims to streamline the traditionally separate OLTP and OLAP systems, reducing manual data syncing efforts. #Databricks #AI #Mooncake #DataAnalytics #Lakebase
Joab Jackson

Domains must now be managed at the team level

2025-12-04 13:00
🔄 Important Update for Domain Management! Domains must now be managed at the team level instead of the account level. This change aims to enhance access control and collaboration while simplifying billing processes. Existing account-level domains will continue to function normally. However, any modifications will require transferring the domain to a designated team. When managing your account-level domain, you will now be prompted to select a team for the transfer. All related project...
Source: Vercel Blog
Rhys Sullivan

The Deprecation of K1 Mode

2025-12-04 09:41
🚀 Exciting news for IntelliJ IDEA users! With the 2025.3 release, K1 mode is officially deprecated as we focus on K2 mode. This transition improves performance, stability, and future Kotlin support. Adoption rates are high, with 98.6% of Ultimate and 99.5% of Community Edition users already using K2 mode. For those still on K1, options remain available for re-enabling. Read more about K2 mode's development and performance improvements in our latest blog. #IntelliJ #Kotlin #K2Mode...
Maria Kosukhina

Adding Async Streaming to Postgres Foreign Data Wrappers

2025-12-04 07:00
🚀 Supabase Wrappers is enhancing Postgres Foreign Data Wrappers (FDWs) by introducing asynchronous streaming! This update allows more efficient, memory-safe queries for large analytical workloads. Traditionally, FDWs operated synchronously, fetching all results or fixed-size batches, which could block the Postgres backend. With this new approach, users can expect improved performance in querying external databases. #Postgres #DataStreaming #Supabase #ForeignDataWrappers #DatabasePerformance

Build "Sign in with Your App" using Supabase Auth

2025-12-04 07:00
🚀 Supabase Auth has expanded its capabilities! It's now equipped with OAuth 2.1 and OpenID Connect, allowing your project to serve as an identity provider. This means AI agents can authenticate through your app using the Model Context Protocol (MCP), enhancing security and integration. Developers can now build on your platform easily, enabling secure API access and enterprise single sign-on. Explore the new possibilities with Supabase! 🔑✨ #Supabase #Auth #IdentityProvider #APISecurity #OAuth

Simplify container management with Bitbucket Packages (now GA)

2025-12-04 05:28
🚀 Bitbucket Packages is now generally available! This native container registry integrates seamlessly into Bitbucket Cloud. Manage your source code, CI/CD pipelines, and container images all in one place, reducing context switching and simplifying permission management. 💰 Note: A paid pricing model will take effect in your next billing cycle. Review and remove any unused images to control costs. #Bitbucket #ContainerManagement #DevOps #Atlassian #CloudComputing
Ash Moosa

Simplify container management with Bitbucket Packages (now GA)

2025-12-04 05:28
🚀 Bitbucket Packages is now generally available, offering a native container registry within Bitbucket Cloud! This feature allows teams to manage source code, CI/CD pipelines, and container images all in one place, enhancing productivity by reducing context switching. Bitbucket Packages will adopt a paid pricing model starting with your next billing cycle. It’s advisable for beta users to review and remove any unused images to manage costs effectively. Key benefits include streamlined...
Ash Moosa

Announcing powerful upgrades & a new pricing model for self-hosted runners

2025-12-04 02:48
🚀 Exciting news from Bitbucket! The release of Bitbucket Pipelines self-hosted runners v5.0 brings significant upgrades. Users can now customize resource allocations, manage storage with Docker volume mounts, and store build data in their private cloud. 🔒 A new paid licensing model will support future enhancements and enterprise-grade experiences while existing runners remain free until June 3, 2026. 💡 Upcoming features include ephemeral macOS runners for scalable iOS/macOS development. Join...
Edmund Munday

Building Holiday Magic: Twilio Powers Real-Time Calls for ‘NORAD Tracks Santa’

2025-12-04 00:00
This December, Twilio joins the NORAD Tracks Santa program, a cherished tradition for over 70 years. 🎅✨ Twilio's WebRTC technology will facilitate real-time calls, connecting children worldwide with volunteers tracking Santa's journey on Christmas Eve. Exciting times ahead as we enhance holiday joy! 🎄📞 #Twilio #NORADTracksSanta #HolidayMagic #WebRTC #SantaTracker
Kevin Wagner

GPT-5.1-Codex-Max now available in AI Gateway and Agent Runners

2025-12-04 00:00
🚀 OpenAI has launched the GPT-5.1-Codex-Max model via Netlify's AI Gateway and Agent Runners, requiring no configuration. Developers can integrate the OpenAI SDK directly into their Netlify Functions without managing API keys. The AI Gateway simplifies this process. This model supports various functions and offers automatic access to Netlify's infrastructure. Explore more in the AI Gateway and Agent Runners documentation. #OpenAI #Netlify #AI #GPT5 #TechNews
Source: Netlify Blog

Unity 6.3 LTS is now available

2025-12-04 00:00
🚀 Exciting news! Unity 6.3 LTS is now available, offering two years of dedicated support and consistent updates. This release simplifies cross-platform development with the new Platform Toolkit, enhancing accessibility for developers and players alike. Key features include support for Nintendo Switch™ 2, Android XR capabilities, and improved performance and stability. For full details, check the release notes! #Unity3D #GameDev #UnityLTS #GameDevelopment #Unity6.3
Source: Unity Blog

What Adobe’s New AI Assistant Can Teach Frontend Developers

2025-12-03 23:00
Adobe's new AI Assistant aims to enhance accessibility in web and application development, according to Navya Agarwal, a senior software engineer at Adobe. She emphasizes that accessibility is not just altruism but a pathway to market growth. The AI Assistant, part of Adobe Express, integrates accessibility features from the start, addressing needs for users with cognitive disabilities, such as ADHD and dyslexia. It offers tailored design adjustments to improve user experience. This...
Loraine Lawson

AWS Makes It Easier To Customize AI Models in Bedrock and SageMaker Without a Ph.D.

2025-12-03 19:30
🚀 AWS has introduced Nova Forge, simplifying the customization of its Nova large language models for enterprises. With new options in Amazon Bedrock and SageMaker, developers can now tailor AI models without needing advanced machine learning expertise. SageMaker offers a guided experience for customization, while Bedrock focuses on Reinforcement Fine Tuning for better task performance. No Ph.D. required! 🌟 #AWS #AI #MachineLearning #SageMaker #NovaForge
Frederic Lardinois

The Forge Fundamentals Certificate Is Now Available

2025-12-03 17:16
🎉 Exciting news! The Forge Fundamentals certificate is now available for free! This official credential validates your skills in Forge development. 📚 The exam tests your knowledge of Forge architecture, app deployment, implementation patterns, and troubleshooting. It consists of 20 questions, with a passing score of 70%. 🔗 Joining this community connects you with other credentialed developers and enhances your professional profile. #ForgeFundamentals #Atlassian #DeveloperCommunity...
Corey L Weathers

Powering the agentic ecosystem with Atlassian Rovo MCP Server

2025-12-03 17:00
🚀 Exciting news for Atlassian users! The new Atlassian Rovo MCP Connector for ChatGPT is now available. This integration allows users to access live updates from Jira and Confluence directly in ChatGPT, enhancing productivity and collaboration. Key features include: - Instant summaries of Jira work items and Confluence pages - Ability to create Jira issues directly within ChatGPT - Automated multi-step actions for efficiency This tool aims to streamline workflows and improve team...
dkoudsi@atlassian.com

Amazon Bedrock adds reinforcement fine-tuning simplifying how developers build smarter, more accurate AI models

2025-12-03 16:08
🚀 Amazon Bedrock introduces reinforcement fine-tuning, enhancing AI model customization. This new feature allows developers to improve model accuracy by an average of 66% without needing large labeled datasets. It simplifies the process, making advanced AI accessible to more developers. Key benefits include better performance, ease of use, and enhanced security within the AWS environment. #AmazonBedrock #AI #MachineLearning #Innovation #TechNews
Source: AWS Blog
Donnie Prakoso

New serverless customization in Amazon SageMaker AI accelerates model fine-tuning

2025-12-03 16:08
🚀 Exciting news for AI developers! Amazon SageMaker AI introduces serverless customization to accelerate model fine-tuning. With new features, you can recover from failures quickly and automatically scale resources. This allows customization of popular models like Nova and GPT-OSS in days instead of months! Simply select a model and choose your customization technique—all with a user-friendly interface. Explore advanced settings and track experiments seamlessly. Get started today in Amazon...
Source: AWS Blog
Channy Yun (윤석찬)

Introducing checkpointless and elastic training on Amazon SageMaker HyperPod

2025-12-03 16:07
🚀 Exciting updates from Amazon SageMaker HyperPod! Introducing **checkpointless training** and **elastic training** features to enhance AI model development. Checkpointless training allows for quick recovery from failures, reducing downtime from hours to minutes. This means teams can focus on improving model performance without lengthy interruptions. Elastic training automatically scales workloads based on resource availability, optimizing cluster utilization and saving engineering time....
Source: AWS Blog
Channy Yun (윤석찬)

Helm 4: What’s New in the Open Source Kubernetes Package Manager?

2025-12-03 16:00
🚀 Helm 4 has officially launched at KubeCon + CloudNativeCon North America, marking the first major update in six years! Originally created as a hackathon project called Kate's Place, Helm has evolved into a vital open-source package manager for Kubernetes. The new version brings improved logging, dependency management, and WebAssembly plugins for enhanced portability. Learn more about its journey and features! 🌐 #Kubernetes #Helm4 #OpenSource #KubeCon #CloudNative
Heather Joslyn

Confidence Through Clarity: Managing the Modern Network

2025-12-03 15:46
Unlock the benefits of network management with Cisco + Auvik! A recent webinar highlighted how their partnership simplifies network visibility and lifecycle control. Automated tools tackle challenges like complexity and device management, improving operational planning. Missed the live session? You can still request a demo through the Cisco Marketplace for tailored insights. 🌐📊✨ #NetworkManagement #Cisco #Auvik #ITTrends #TechInsights
Ana Nennig

New npm package for automatic recovery of broken streaming markdown

2025-12-03 13:00
🚀 Introducing Remend, a new npm package for seamless Markdown recovery! This standalone library enhances any application by intelligently handling incomplete Markdown syntax. It detects and completes unterminated blocks, ensuring clean rendering during streaming. Remend works with any Markdown renderer and has proven effective in production AI applications. Get started with Remend today! #Markdown #WebDevelopment #AI #npm #Remend
Source: Vercel Blog
Hayden Bleasel

Nova 2 Lite now available on Vercel AI Gateway

2025-12-03 13:00
📢 Exciting news! Amazon's Nova 2 Lite is now accessible through Vercel's AI Gateway without needing other provider accounts. This reasoning model can handle text, images, and videos to generate text outputs. To use it, simply set the model to amazon/nova-2-lite in the AI SDK. Note that reasoning is disabled by default but can be enabled with specific settings. The AI Gateway offers a unified API for model calls, usage tracking, and performance optimizations. Explore more in the model...
Source: Vercel Blog
Jerilyn Zheng

Vercel Agent can now install Web Analytics and Speed insights for you

2025-12-03 13:00
🚀 Vercel Agent now simplifies adding Web Analytics and Speed Insights to your projects using AI! To get started, navigate to your Vercel dashboard, enable the feature, and let the agent handle the installation. It analyzes your project, installs the necessary packages, and creates a pull request for your review. Once merged, tracking begins automatically. This feature is in Public Beta for all teams. 🌐✨ #WebAnalytics #SpeedInsights #Vercel #AI #TechUpdates
Source: Vercel Blog
Casey Gowrie

Introducing Seven New Email Templates for Supabase Auth

2025-12-03 07:00
🔔 Exciting news for Supabase users! We have launched 7 new email notification templates for Supabase Auth. These templates aim to enhance account security by informing users about critical changes on their accounts. Users will receive alerts for actions such as password changes, email updates, and linked identities. Stay informed and safe! 🔒✉️ #Supabase #AccountSecurity #EmailNotifications #UserSafety #TechUpdates

The new Supabase power for Kiro

2025-12-03 07:00
🚀 Exciting news for developers! The latest update brings Supabase powers to Amazon's Kiro IDE, enhancing your ability to build full-stack applications efficiently. With Kiro's deep integration, you get access to your Supabase database schema, edge functions practices, and security guidelines for Row Level Security (RLS). This update allows for streamlined management of database schemas and offers one-click access to essential development tasks. Two new powers are available: one for hosted...

Stack Overflow Puts Community First in New AI Search Tool

2025-12-03 01:30
🚀 Stack Overflow has launched AI Assist, a new AI search tool that emphasizes human-verified community knowledge. This tool uses a hybrid approach, prioritizing user-generated content and ensuring accurate results with proper attribution. It aims to provide developers with reliable answers quickly and for free. Over 250,000 technologists are already using AI Assist for various technical inquiries. Explore more at stackoverflow.com/ai-assist! 🔍💡 #StackOverflow #AI #TechCommunity #Innovation...
Darryl K. Taft

Bitbucket’s new look: user experience and navigation updates coming soon

2025-12-03 00:36
🚀 Bitbucket is getting a fresh new look! As part of Atlassian's visual system journey, updates will enhance user experience and navigation. 🔍 Key changes include streamlined navigation for quicker access to repositories, pull requests, and pipelines. This aims to reduce context switching and improve clarity. 📅 The new features will progressively roll out over the next few weeks, with a full navigation update expected in early 2026. 📣 Share this news with your team and prepare for the upcoming...
Markus te Grotenhuis

Explore the updates to the Game Animation Sample Project in UE 5.7

2025-12-03 00:00
🚀 Exciting updates are here for the Game Animation Sample Project (GASP) in UE 5.7! New features include the Mover plugin, Smart object setups, a slide feature, and a fresh locomotion style. These additions enhance the project and expand creative possibilities. Explore the latest developments and elevate your game animation experience! 🎮✨ #GameDevelopment #UE5 #Animation #GASP #Updates

With Nova Forge, AWS Makes Building Custom AI Models Easy

2025-12-02 21:09
🚀 AWS has launched Nova Forge at the annual re:Invent conference, introducing new versions of their Nova models. This service allows businesses to customize AI models with proprietary data. 💡 Users can access pre-, mid-, and post-trained Nova models while fine-tuning responses through reinforcement learning. 🔒 AWS emphasizes safety with its responsible AI toolkit. Companies like Reddit are already seeing improved efficiency in content moderation. #AWS #AI #NovaForge #MachineLearning #TechNews
Frederic Lardinois

AWS Frontier Agents Handle Code, Security, Ops Without You

2025-12-02 20:00
🚀 Amazon Web Services has introduced three autonomous AI agents for the software development lifecycle, announced at the AWS re:Invent conference. These frontier agents—Kiro, AWS Security Agent, and AWS DevOps Agent—can operate independently, managing multiple tasks without constant human oversight. They aim to enhance productivity by allowing developers to focus on broader goals rather than micromanaging every detail. As these agents learn from team dynamics, they can submit pull requests...
Darryl K. Taft

NVIDIA-Accelerated Mistral 3 Open Models Deliver Efficiency, Accuracy at Any Scale

2025-12-02 18:10
🚀 Exciting news in AI! The Mistral 3 open model family has been launched, offering enhanced accuracy and efficiency for developers and enterprises. This suite features a large multimodal and multilingual model with 675B parameters, alongside smaller high-performance models (3B, 8B, 14B) for diverse applications. All models are trained on NVIDIA Hopper GPUs and available via Mistral AI on Hugging Face, providing versatile deployment options. #AI #MachineLearning #NVIDIA #Mistral3 #TechInnovation
Anu Srivastava

Introducing Stack Overflow AI Assist—a tool for the modern developer

2025-12-02 18:00
🚀 Introducing AI Assist at Stack Overflow! This innovative tool transforms how developers access knowledge. By combining human-verified answers with generative AI, users can find solutions faster and with less hassle. Our team engaged with users to create an experience that eases barriers and enhances learning, ensuring everyone feels at home in our community. #StackOverflow #AI #DeveloperTools #KnowledgeSharing #TechInnovation
Phoebe Sajor

When GenAI Meets Personalization: Powering DoorDash’s next-generation homepage experience

2025-12-02 17:43
🚀 Exciting updates from DoorDash! We've launched a GenAI-powered personalized carousel system on our homepage, enhancing your shopping experience. This system tailors recommendations using large language models, creating unique carousels for each user based on their preferences. By analyzing customer data, we're reducing search challenges and helping you make quicker, informed decisions. This means more relevant options and a smoother journey on our platform! #DoorDash #GenAI #Personalization...
Yuxiang Wang

AWS Updates Its Nova Models To Compete With Google, Anthropic and OpenAI

2025-12-02 17:15
🚀 AWS has unveiled its latest Nova models at the re:Invent conference in Las Vegas. The Nova 2 Lite, Pro, Sonic, and the new Omni model can process text, audio, images, and video. These updates aim to enhance AWS's competitiveness against leading models from Google, OpenAI, and Anthropic. Pricing starts at $0.3 per million tokens for the Lite model. The Nova 2 Lite is designed for customer service and workflow automation, while the Nova Pro focuses on accuracy and reasoning across various...
Frederic Lardinois

AWS, Google Build a Multicloud Bridge

2025-12-02 17:00
🌐 AWS and Google Cloud have teamed up to tackle multicloud connectivity challenges. They developed a standard for Layer 3 connectivity, allowing seamless networking between their cloud services. This initiative aims to ease the integration of cloud operations and alleviate concerns about cloud lock-in. The open-source Connection Coordinator API will help provision dedicated bandwidth between the two providers. AWS has implemented this in their AWS Interconnect, while Google Cloud has its...
Joab Jackson