Articles from Source: The-New-Stack

Introduction to Observability

2025-09-27 12:05
🌐 Observability is a key concept in IT and DevOps that goes beyond just monitoring and debugging. It automates analysis and decision-making, enhancing operational processes. Think of it like baking a cake where every ingredient and condition is monitored for success. While metrics, logs, and traces are integral, they should be processed together for a complete view of system behavior. #Observability #DevOps #ITOperations #OpenTelemetry #TechInsights
Source: The New Stack
B. Cameron Gain

Microsoft AI Agents Automate Enterprise Java and .NET Migrations

2025-09-26 23:00
Microsoft has launched AI-powered development tools to simplify and speed up the migration of legacy enterprise applications like .NET and Java. 🖥️🚀 These tools aim to cut migration time from months to days, addressing the challenge of outdated application portfolios. Over 37% of enterprise applications need modernization, which can lead to security risks and technical debt. 🔄🔒 The AI agents help assess application dependencies and streamline workflows, making the migration process more...
Source: The New Stack
Darryl K. Taft

Python Dataclasses: A Complete Guide to Boilerplate‑Free Objects

2025-09-26 22:00
Discover the power of Python dataclasses! 🐍 Dataclasses provide a structured way to store related values with minimal boilerplate code. Unlike variables, dictionaries, and regular classes, dataclasses streamline object creation and enhance code maintainability. With the @dataclass decorator, enjoy built-in methods like __init__, __repr__, and __eq__, making your code cleaner and more efficient. Learn how to leverage dataclasses for better data organization! 📊 #Python #Dataclasses #Programming...
Source: The New Stack
Jessica Wachtel

Sentry Founder: AI Patch Generation Is ‘Awful’ Right Now

2025-09-26 21:00
David Cramer, founder of Sentry, expresses skepticism about generative AI's current effectiveness in software development. He believes that while GenAI tools are not yet reliable for production use, there is potential in AI-based products. Sentry's recent launch, Seer, utilizes GenAI to enhance debugging and root cause analysis. Cramer highlights how large language models (LLMs) can summarize data and identify problems like experienced engineers. However, he emphasizes that AI still needs a...
Source: The New Stack
Heather Joslyn

Vibe Coding: When AI Writes the Code, Who Secures It?

2025-09-26 20:00
Exploring "vibe coding" introduces exciting possibilities for creators using AI tools like ChatGPT and Copilot. This approach encourages innovation but raises critical security concerns. Experts, including prominent figures from the open-source community, emphasize the need for human oversight in AI-generated code. While AI can enhance rapid prototyping, risks remain without proper guidance. As we embrace AI in coding, it’s essential to balance enthusiasm with experience and maintain security...
Source: The New Stack
Crystal Morin

Clarifai’s AI Engine Cuts Costs Without Performance Hit

2025-09-26 19:00
Clarifai has introduced a new tool aimed at reducing the costs associated with AI models. 💡 The Clarifai Reasoning Engine optimizes performance without sacrificing quality. CEO Matt Zeiler explains that this engine can make models two times faster at only 40% of the typical cost. 🚀 Organizations can easily integrate their custom AI models using this platform, streamlining their processes. #AI #CostEfficiency #Clarifai #TechInnovation #MachineLearning
Source: The New Stack
Loraine Lawson

How To Build AI Agents With Lifelong Learning

2025-09-26 18:00
AI is evolving beyond the "set it and forget it" model. Traditional language models like GPT-4 can't learn from new experiences after deployment. Continual learning agents address this gap, improving over time by remembering conversations and adapting to customer needs. They enhance functions in sales, research, and customer service. This article offers a step-by-step guide to build such adaptive agents using existing technologies. Learn how to create a conversational agent that evolves with...
Source: The New Stack
Oladimeji Sowole

Why pgEdge ‘Ripped the Band-Aid Off’ To Go Totally Open Source

2025-09-26 17:00
pgEdge has made a significant shift by going fully open source, transitioning its Spock multimaster logical replication extension and other tools to an OSI-approved PostgreSQL License. This decision came in response to potential customers wanting true open-source solutions. CEO Phillip Merrick indicated that their previous licensing structure limited engagement with the Postgres community. Merrick acknowledged concerns about competition but emphasized the need for clarity in the open-source...
Source: The New Stack
Susan Hall

How Vite Became the Backbone of Modern Frontend Frameworks

2025-09-26 16:00
🚀 Vite, developed by Evan You, is now the leading build tool for modern frontend frameworks like Vue, SvelteKit, and Astro, among others. 📅 A documentary detailing Vite's journey will be released on Oct. 9. 🔍 The tool emerged from You's desire for a faster, lighter development experience, leveraging the capabilities of ES Modules, which became widely supported in recent years. 💡 Vite enables real-time compilation of modules, eliminating the need for traditional bundling, and features hot...
Source: The New Stack
Richard MacManus

CISOs: Prepare for Software’s Agentic Future Today

2025-09-26 15:00
📊 A recent GitLab survey reveals that 89% of executives foresee agentic AI becoming the key standard in software development within three years. However, 85% acknowledge the significant security challenges it presents. 🔒 CISOs face the tough task of adopting AI while minimizing its risks. With 91% planning to increase AI spending, the need for a robust governance model is critical. Yet, 47% lack regulatory-aligned governance, highlighting the urgency for action. ⚖️ To tackle these challenges,...
Source: The New Stack
Josh Lemos

Why Rancher’s Founders Pivoted From Kubernetes to Agentic AI

2025-09-26 14:00
Rancher’s founders have shifted their focus from Kubernetes to agentic AI, marking a pivotal change in enterprise tech. 🚀 As generative AI evolves into autonomous agents, organizations face new challenges in governance, security, and scalability. The rise of “shadow AI” complicates this landscape, necessitating a dedicated control plane. Obot.ai, founded by this experienced team, aims to provide crucial governance as the standard Model Context Protocol (MCP) emerges. 🌐 #AgenticAI...
Source: The New Stack
Janakiram MSV

Project infragraph: IBM’s Real-Time Model for Infrastructure Assets

2025-09-26 11:00
🚀 IBM has introduced Project infragraph at HashiConf 2025, aiming to improve infrastructure management. This initiative seeks to unify IT assets, security, and applications under one data model, enhancing visibility and compliance. The project will enter a private Beta phase in December and integrates with platforms like Red Hat Ansible and IBM watsonx Orchestrate. Stay tuned for advancements in real-time infrastructure and AI-driven workflows! #IBM #ProjectInfragraph...
Source: The New Stack
Joab Jackson

Okta Wants To Secure Your AI Agents, Too

2025-09-25 21:00
At the Oktane user conference, Okta announced its initiative to secure AI agents as key entities within its platform. CEO Todd McKinnon emphasized the importance of identity management for AI agent security, advocating for a comprehensive approach to data access. Okta's new product, "Okta for AI Agents," aims to standardize secure connections via the Cross App Access protocol. This will enhance visibility and control for IT and security teams. The rollout will include tools for discovering...
Source: The New Stack
Frederic Lardinois

JavaScript Gets Supply Chain Security With Chainguard Libraries

2025-09-25 19:00
🚀 Chainguard has launched Chainguard Libraries for JavaScript, a new initiative aimed at enhancing supply chain security by providing thousands of JavaScript dependencies rebuilt from source. This move follows recent malware attacks on popular npm packages that have raised concerns within the developer community. Companies are taking precautions, pausing development to assess vulnerabilities. Chainguard’s unique approach involves rebuilding source code from GitHub, bypassing potential risks...
Source: The New Stack
Darryl K. Taft

How To Use RAG-Powered AI for Safer Legacy Code Maintenance

2025-09-25 17:00
Navigating legacy code can be challenging for enterprises. As generative AI (GenAI) quietly integrated into coding practices, developers found it helpful for routine tasks. However, maintaining legacy systems remains a significant concern. Outdated code and lack of documentation complicate updates and integration. Companies are turning to agentic AI, like Tabnine's RAG-powered Enterprise Context Engine, to enhance code understanding and accuracy. By analyzing and adapting to changes in the...
Source: The New Stack
David Eastman

Microsoft Launches a Combined Marketplace for Cloud Solutions, AI Apps and Agents

2025-09-25 16:00
🚀 Microsoft has launched the "Microsoft Marketplace," combining the Azure Marketplace and Microsoft AppSource. This unified platform aims to simplify the procurement of cloud solutions, AI apps, and agents for businesses. With over 3,000 AI tools available, it responds to the growing demand for trusted applications. Microsoft emphasizes the need for speed, simplicity, and confidence as companies adopt AI. Additionally, partners can integrate the marketplace into their own offerings, enhancing...
Source: The New Stack
Frederic Lardinois

Consistency at Scale: Unifying Temporal and YugabyteDB

2025-09-25 15:00
Manetu has unified Temporal and YugabyteDB, creating a robust data platform that enhances AI reliability and governance. This open-source integration simplifies operations and strengthens trust at scale. By merging orchestration and persistence, Manetu addresses critical infrastructure challenges, ensuring workflows execute smoothly even under stress. This advancement not only boosts performance but also reinforces customer confidence in data integrity and governance. #DataIntegration #AI...
Source: The New Stack
Greg Haskins

Why Your App’s Biggest Performance Bottleneck Might Be SSL/TLS

2025-09-25 13:00
Is your app's performance lagging despite optimizations? The culprit may be SSL/TLS. 🔒 While we often see it as a security feature, SSL/TLS can be a significant performance bottleneck. Each secure connection requires CPU-intensive handshakes that can compete with your app's logic. Understanding the negotiation process is essential. It involves greeting, certificate exchange, and key generation—each step adds overhead. Stay aware of these factors to improve your app's efficiency! 💻...
Source: The New Stack
Ron Northcutt

The Cross App Access Protocol Makes AI Agents Enterprise-Ready

2025-09-25 12:00
Enterprises are integrating AI agents, but governance systems are lacking. 🤖💼 The new Cross App Access (XAA) protocol aims to enhance security by allowing agents to access data within established identity management frameworks. Supported by companies like Okta and AWS, XAA gives security teams better control and oversight. 🛡️🔑 By complementing existing protocols, it simplifies access while ensuring agents are authenticated and authorized. This could improve the user experience and reduce...
Source: The New Stack
Frederic Lardinois

How To Deploy a Full-Stack, Containerized Network Infrastructure Visualizer

2025-09-24 20:00
Struggling to visualize your busy network? 🖥️ The article discusses deploying Atlas, a full-stack, containerized network infrastructure visualizer. It simplifies monitoring by scanning Docker containers and local hosts, gathering essential data like IP addresses and open ports. With a straightforward setup, Atlas can be up and running in just two minutes! 🚀 Learn more about its features and how it can help you manage your network effectively. #NetworkManagement #Atlas #TechTools...
Source: The New Stack
Jack Wallen

Top 10 Kubernetes Deployment Errors: Causes and Fixes (And Tips)

2025-09-24 18:00
Kubernetes deployment failures can be challenging to diagnose. Misconfigurations are often the main cause, impacting security and stability. The article outlines 10 common errors, such as CrashLoopBackOff and YAML issues, along with strategies to troubleshoot them effectively. Understanding why these errors occur can help prevent future issues. Key reasons include misconfigured YAML files, incorrect image names, and resource constraints. Stay informed to ensure smoother deployments! 🚀🔧...
Source: The New Stack
Sunny Yadav

Google’s Data Commons Gives AI Agents Access to a Vast Trove Of Stats

2025-09-24 17:30
📊 Google’s Data Commons is enhancing data accessibility for AI agents by collecting public datasets across various topics. This initiative helps integrate real-world statistics into applications, improving the reliability of responses from AI models. 🌐 With the launch of the Model Context Protocol server, developers can easily access this data without direct API interaction. Google provides tutorials for integration with various tools, making it simpler to generate insightful reports. 🤝 A...
Source: The New Stack
Frederic Lardinois

The Case for Microfrontends and Moving Beyond One Framework

2025-09-24 16:00
Building large applications today presents unique challenges, especially for the frontend. 🖥️ The article discusses the shift from monolithic frontends to microfrontends. This approach allows teams to create independently deployable slices of code, improving deployment safety and reducing bottlenecks. 🚀 However, this model introduces complexity in shared state and design consistency. Organizations can benefit from flexibility in choosing frameworks without being locked into one, but they must...
Source: The New Stack
Alexander T. Williams

A Guide To Fluent Bit Processors for Conditional Log Processing

2025-09-24 15:00
Unlock the power of Fluent Bit with its latest processors for conditional log processing! 📊 Fluent Bit, an open-source data collector, has introduced processors that modify telemetry data effectively. With version 4, you can now apply conditional processing to logs based on specific field values. Processors differ from filters in execution timing and performance, enhancing efficiency. For detailed setup, check the YAML configuration requirements. #FluentBit #LogProcessing #OpenSource...
Source: The New Stack
Sharad Regoti

All Infrastructure Is AI Infrastructure

2025-09-24 14:00
In the near future, all technology infrastructure will transform into AI infrastructure. As enterprises adopt generative AI and autonomous agents, every layer of the tech stack will be reshaped to support AI workloads. This shift mirrors the transition from physical data centers to cloud computing. Conventional applications may soon be replaced by AI-driven systems designed and adapted in real-time. AI infrastructure demands a detailed understanding of hardware, reversing the trend of...
Source: The New Stack
Shaun O'Meara

AI Has Won: Google’s DORA Study Shows Universal Dev Adoption

2025-09-23 23:00
🚀 A significant shift in software development is underway, as Google's DORA report reveals that 90% of tech professionals are now using AI tools, up 14% from last year. Developers are spending about two hours daily with AI, enhancing productivity for over 80% of users. Yet, trust in AI output varies, with 30% expressing low confidence in its results. Human oversight remains crucial, even at Google, where 30% of new code is AI-generated but reviewed by engineers. #AI #SoftwareDevelopment...
Source: The New Stack
Darryl K. Taft

Open Source Turmoil: RubyGems Maintainers Kicked Off GitHub

2025-09-23 17:31
🚨 The RubyGems community is in turmoil after maintainers were removed from the GitHub repository. Earlier this month, Ruby Central replaced all maintainers, sparking complaints of a hostile takeover. An anonymous maintainer initially changed permissions, leading to confusion and backlash. Ruby Central later cited security concerns as the reason for their actions, claiming a fiduciary duty to protect the ecosystem. #OpenSource #RubyGems #GitHub #Community #TechNews
Source: The New Stack
Loraine Lawson

Snowflake, Salesforce Launch New Standard To Unify Data for AI

2025-09-23 16:00
🌐 Snowflake and Salesforce have launched the Open Semantic Interchange (OSI), a vendor-neutral initiative aimed at standardizing data for AI. This framework intends to unify contextual information for structured and unstructured data, allowing organizations to collaborate effectively. Key players include dbt Labs, BlackRock, Mistral AI, and RelationalAI. The OSI will not define metrics but will provide a format for users to do so. #DataStandardization #AI #BusinessIntelligence...
Source: The New Stack
Joab Jackson

GPT-5’s Enhanced Reasoning Comes With a Steep Hidden Cost

2025-09-23 15:00
🚀 The arrival of GPT-5 marks a major advancement in AI-driven code generation. It's proficient in tackling complex programming tasks effectively. 🔍 However, a recent analysis by Sonar highlights a hidden cost: the quality and maintainability of the code it generates may suffer. While it accelerates development, it can lead to increased technical debt. 🛠️ GPT-5 produces more lines of complex code and introduces more issues compared to other models. Its strengths in security are overshadowed by...
Source: The New Stack
Prasenjit A. Sarkar

How To Enhance Productivity With DORA Metrics

2025-09-23 14:00
Enhancing productivity in software development involves more than just writing clean code. The article discusses DORA metrics, which are key performance indicators that help teams optimize workflows and track business outcomes. These metrics provide insights into deployment frequency and lead time, allowing teams to identify inefficiencies and improve speed. Regular monitoring can lead to better deployment stability and clearer performance tracking. DORA metrics also enhance visibility across...
Source: The New Stack
Susan Hall

Why You Can’t Debug a Running Quantum Computer Program

2025-09-23 13:00
🔍 Debugging quantum computer programs poses unique challenges. Unlike traditional programming, errors cannot be fixed once the code is running due to high costs and limited capabilities of quantum hardware. 💻 Mariia Mykhailova from PsiQuantum highlights the importance of thorough pre-execution testing and outlines a structured workflow for quantum software development. 📊 She emphasizes that not all tasks are suitable for quantum computing, particularly those involving large data sets. The...
Source: The New Stack
Joab Jackson

TikTok’s Ex-Algorithm Chief Launches Verdent AI Coding Tool

2025-09-23 13:00
🚀 Exciting news in the tech world! Zhijie Chen, former Head of Algorithms at TikTok, has launched Verdent, an AI coding tool designed to transform software development. With two versions available—a Visual Studio Code plugin and a desktop app—Verdent aims to enhance coding efficiency. Chen emphasizes that Verdent stands out with its full-stack AI engine, customizable workflows, and multitasking capabilities, making it unique compared to existing tools. Discover how lessons from TikTok's...
Source: The New Stack
Richard MacManus

How to Fix Performance Issues Error Monitoring Can’t See

2025-09-22 16:16
🚀 Users report "buggy" experiences despite high crash-free rates. Why? Performance issues like slow startup times and memory leaks often go unnoticed by error monitoring. Real User Monitoring (RUM) is essential for identifying these hidden performance bottlenecks. It captures real-world app behavior under various conditions. Key performance problems include cold startup delays and app bloat, which can affect user experience. Tools to monitor and address these issues can help maintain optimal...
Source: The New Stack
Bob Nelson

How We Cut Telemetry Queries to Under 10 Milliseconds

2025-09-22 15:00
🚀 We developed a telemetry pipeline that processes over 5,400 data points per second with response times under 10 milliseconds. 📊 By utilizing techniques from flight simulator data, we improved query performance significantly. Traditional queries took over 30 seconds, but with caching and batching, we reduced this to less than 10ms. ⚙️ Key strategies included implementing Last Value Cache and batch writing, resulting in thousands of metrics processed with no data loss. #Telemetry...
Source: The New Stack
Heather Downing

Is Oreon Linux a True Red Hat Enterprise Alternative?

2025-09-21 14:00
Is Oreon Linux a viable alternative to Red Hat Enterprise Linux? This article explores the landscape of Linux distributions for enterprise users, highlighting options like RHEL and SUSE. Oreon Linux emerges as a user-friendly choice, based on AlmaLinux, offering stability and security. Oreon aims to balance simplicity with enterprise needs, featuring minimal pre-installed software and extended support until 2035. Discover how Oreon could serve both businesses and individual users seeking a...
Source: The New Stack
Jack Wallen

Java Language Architect Brian Goetz on How Java Could Evolve

2025-09-21 13:00
Brian Goetz, Java's language architect, recently shared insights on Java's future at the JVM Summit. He discussed a vision for a more extensible and growable Java, inspired by past programming language philosophies. Goetz emphasized that this is not about immediate features but rather a long-term evolution of the language. He reflected on the importance of making Java adaptable for developers. A key point was the introduction of 'Witnesses', a new concept aimed at enhancing Java's...
Source: The New Stack
David Cassel

Wasm 3.0 Offers New Way to Handle JavaScript Strings

2025-09-20 14:14
🚀 Wasm version 3 has been released by the W3C Community Group, marking a significant update with features developed over six to eight years. A key enhancement for frontend developers is the new method for handling JavaScript strings. Wasm now allows direct access and manipulation of these strings, improving support for high-level languages like Java and Kotlin. Additional updates include support for 64-bit address space, multiple memories within a single module, and a new garbage collection...
Source: The New Stack
Loraine Lawson

AWS Kiro: Testing an AI IDE with a Spec-Driven Approach

2025-09-20 12:05
🚀 In July, AWS launched Kiro, an AI IDE focused on spec-driven development rather than prompt-based interactions. 🔍 Kiro emphasizes generating user stories, creating technical designs, and breaking tasks into trackable sequences. This structured approach caters to agile teams. 💻 After a waitlist, users can access Kiro, which resembles Visual Code, to streamline their coding projects. #AWS #AIIDE #Kiro #SoftwareDevelopment #TechNews
Source: The New Stack
David Eastman

Microsoft Goes All-in on Rust for Core Infrastructure and Much More

2025-09-20 00:00
Microsoft's Azure CTO, Mark Russinovich, recently highlighted the company's deep integration of Rust in its infrastructure at RustConf 2025. He showcased projects like Hyperlight, which spins up virtual machines in just 1.5 milliseconds, enhancing performance and isolation for untrusted code. This system, originally a C# prototype, was rewritten in Rust to meet critical performance needs. Additionally, Microsoft introduced a Cargo plugin for MSBuild, allowing Rust modules to be integrated...
Source: The New Stack
Darryl K. Taft

Why Linear Built an API For Agents

2025-09-19 20:15
🚀 Linear has integrated Cursor's AI agent into its project management tool, enhancing how users fix issues and implement design changes. 🤖 This partnership aims to make coding agents essential within Linear, allowing developers to assign tasks to the Cursor agent for background processing. 💡 As developers adapt to using agents, it’s crucial to understand their capabilities and provide clear guidance for optimal results. #Linear #Cursor #ProjectManagement #CodingAgents #TechIntegration
Source: The New Stack
Frederic Lardinois