2026-08-19 07:01
π The Ansible development tools Model Context Protocol (MCP) server enhances AI-driven automation by bridging the gap between describing tasks and executing them. π§ With MCP, AI assistants can directly run commands, manage environments, and generate playbooks tailored to your setup, reducing context-switching and fragmented workflows. π This development enables a more reliable and efficient automation process, allowing developers to focus on coding instead of troubleshooting AI-generated...
Source: Red Hat Developer Blog
Leonardo Gallego
2026-08-19 07:01
Understanding the tokenomics of self-hosted large language models (LLMs) is crucial for platform engineers and MLOps leads. When switching from cloud APIs to self-hosted models, costs shift significantly. The cost per token is calculated as operating costs divided by tokens processed. Key factors include hardware, software, and personnel expenses. To lower costs, consider optimizing usage and spending, such as autoscaling resources and choosing the right models. Monitoring both theoretical...
Source: Red Hat Developer Blog
Trevor Royer
2026-08-18 07:01
π Installing Claude Code plug-ins gives third-party code access to your terminal and files. However, many developers install them without reviewing the source code. The ecosystem lacks centralized vetting, relying on trust in developers and their release pipelines. This post outlines best practices for maintaining security and informs users on how to evaluate plug-ins safely. Key practices include ensuring branch protection, requiring peer reviews, and using continuous integration checks....
Source: Red Hat Developer Blog
Benjamin Kapner
2026-08-18 03:01
π As organizations scale their AI initiatives, the Red Hat OpenShift AI dashboard is addressing key challenges with a Models-as-a-Service (MaaS) approach. π This architecture introduces a Backend-for-Frontend (BFF) to manage complexities like CORS issues, token handling, and API interactions, enhancing user experience and security. π The BFF serves as a single API surface, streamlining data access and authentication while ensuring stable API contracts. #RedHat #OpenShift #AI #ModelsAsAService...
Source: Red Hat Developer Blog
Wen Liang
2026-08-17 13:09
Unlock streamlined OS deployment with Red Hat's Image Builder! π This tool allows you to create customized bootable RHEL images tailored for your needs. You can use it locally on RHEL 9/10 or via the Hybrid Cloud Console. Start by creating an image blueprint, selecting your target environment, and customizing your settings. Once built, deploy your image easily! Learn more about image mode for RHEL today! #RedHat #ImageBuilder #CloudComputing #RHEL #TechTools
Source: Red Hat Developer Blog
Shane McDowell
2026-08-17 07:16
π Exciting news for Kubernetes users! The developer preview of the **Krkn Operator** is now available in Red Hat Advanced Cluster Management. This new tool enables Kubernetes-native chaos engineering from a centralized management hub. π With the Krkn Operator, teams can conduct multicluster chaos testing and assess resilience by running experiments that simulate various failure scenarios, ensuring systems can recover effectively before production issues arise. π§ The **Chaos Studio** feature...
Source: Red Hat Developer Blog
Tullio Sebastiani, Naga Ravi Chaitanya Elluri
2026-08-14 07:01
π Discover ways to enhance your developer experience! The article discusses improvements made in Kubeflow Pipelines (KFP) and Open Data Hub (ODH). Key areas of focus include streamlining CI/CD processes, addressing flaky tests, and simplifying builds. Implementing reusable actions can significantly reduce maintenance efforts. Additionally, fixing root causes of test failures and ensuring secure defaults can enhance reliability. Explore these patterns to elevate your projects!...
Source: Red Hat Developer Blog
Helber Belmiro
2026-08-14 03:01
π Red Hat introduces the new --vm-incident mode for kubevirt/must-gather, simplifying virtual machine (VM) incident investigations. This tool focuses on specific VMs, reducing the time and resources needed for data collection. It captures important context, including host data, VM state, and relevant metrics, all within a defined time window. π The new mode aims to enhance incident reporting, making it easier for investigators and supporting AI-assisted analysis. For more details, check the...
Source: Red Hat Developer Blog
Fabian Deutsch, Simone Tiraboschi
2026-08-13 12:57
Discover how the llm-d model serving routes inference traffic on Amazon EKS! π This article explains the journey of a request from the ingress gateway to vLLM pods, highlighting the role of the endpoint picker (EPP) scheduler. The EPP determines the best pod for each request using real-time routing based on Kubernetes custom resources. For detailed insights, watch the accompanying video! π₯ #Kubernetes #LLM #AI #ModelServing #AmazonEKS
Source: Red Hat Developer Blog
Alexa Griffith
2026-08-13 07:16
Struggling with LLM deployment? Neural Navigator offers a data-driven solution to eliminate guesswork. It transforms natural language requirements into optimized infrastructure using a 4-step workflow: intent extraction, research-backed mapping, multi-criteria ranking, and 1-click deployment. This approach prioritizes service level objectives and benchmarks for efficient production. Learn more about enhancing your LLM deployments with precision! ππ #LLM #NeuralNavigator #DataDriven #AI...
Source: Red Hat Developer Blog
Yuval Luria
2026-08-13 03:01
π The article discusses how to build a DIY pipeline for securing software supply chains amid rising cyber threats. It emphasizes the importance of signing, attesting, and verifying container images. The author implements a home lab using Red Hat tools like MicroShift, Gitea, and Sonatype Nexus for this purpose. Key frameworks such as SLSA and tools like Sigstore and Cosign are explored for enhancing security. Custom configurations and comparisons with Red Hat Advanced Developer Suite...
Source: Red Hat Developer Blog
Kevin Chung
2026-08-12 07:15
π New large language models (LLMs) are emerging rapidly, and the vLLM community is keeping pace with Day 0 support for many. To check if your model is compatible with vLLM, start by exploring the Red Hat validated models program. Models in this program have been tested by Red Hat engineers. You can find the catalog on the RedHatAI Hugging Face page. If your model isn't validated, check its architecture in the config.json file. Ensure it aligns with supported architectures on the vLLM models...
Source: Red Hat Developer Blog
Trevor Royer
2026-08-12 03:01
π Red Hat OpenShift enables automatic cryptographic identities for containers via the zero trust workload identity manager, but challenges remain for workloads in virtual machines (VMs). This tutorial outlines how to bridge this gap using VSOCK and a dedicated in-VM SPIRE agent. By implementing this solution, VM workloads can also obtain short-lived, rotating SPIFFE identities, enhancing security. To learn more about setting up this integration and the benefits of VSOCK, check out the full...
Source: Red Hat Developer Blog
Raushan Kumar Singh
2026-08-11 03:01
Unlock secure automation with HashiCorp Vault and Red Hat Ansible Automation Platform! π This integration uses short-lived OpenID Connect (OIDC) credentials to enhance security and address the "secret zero problem." Learn how to enable the OIDC feature, configure both platforms, and retrieve sensitive data during automation execution. For detailed steps and best practices, check out the full article! π #Ansible #HashiCorpVault #Automation #CyberSecurity #SecretsManagement
Source: Red Hat Developer Blog
Andrew Block
2026-08-10 07:01
π Introducing MiDojo, an open-source framework designed to enhance the security of AI agents in real-world environments. π By applying the "bring your own agent" (BYOA) principle, MiDojo enables effective red-teaming, focusing on system-level security rather than just model-level testing. π It features an interception layer that tests agents against real threats, measuring both security and utility to provide clear insights on performance. Explore MiDojo on GitHub and ensure your AI agent is...
Source: Red Hat Developer Blog
Diego Maniloff
2026-08-10 07:01
π The article discusses how the Red Hat build of Podman Desktop integrates with Project Hummingbird to enhance container security. π‘οΈ Developers can now access hardened base images directly within their local workflow, minimizing unnecessary dependencies and reducing vulnerabilities. π₯οΈ The Hummingbird extension simplifies the process by offering 1-click migrations and a catalog of secure images, making it easier to maintain production-ready containers. Explore the benefits of hardened images...
Source: Red Hat Developer Blog
Axel Stefanini, Stevan Le Meur
2026-08-07 14:56
π Red Hat's IT AI Platforms team recently upgraded OpenShift AI from version 2.25.5 to 3.3.2, a complex task that they accomplished in just 8 weeks using an AI coding assistant. This approach, which typically would take 6 months with multiple engineers, reduced engineering effort by about 60%. The upgrade involved a new architecture and tools to ensure zero data loss and minimal user impact. Learn more about their innovative methodology! #OpenShiftAI #RedHat #AI #TechUpgrade #Innovation
Source: Red Hat Developer Blog
Dasharath Masirkar
2026-08-05 07:01
Deploying AI inference services on Red Hat AI using Amazon EKS involves key Kubernetes resources. The process begins with the setup of gateway and controller pods that handle inference requests. Two model deployments showcase basic and intelligent routing capabilities. The LLMInferenceService YAML configuration is crucial for defining the model, replicas, and GPU resources. Explore how KServe optimizes model serving for large language models with advanced routing features. #Kubernetes #AI...
Source: Red Hat Developer Blog
Alexa Griffith
2026-08-05 03:01
π During my internship at Red Hat, I explored improving hashmap performance in OVN and OVS by analyzing modern hash table designs. 𧩠My research built on a project by Rosemarie O'Riorden, who tested a new hashmap design. We compared its performance with existing implementations, focusing on metrics like insert and search times. π I also investigated Swiss tables, known for their efficiency, and created a new C library called dshmap for better integration. π Benchmarks showed that while some...
Source: Red Hat Developer Blog
Nicholas Hubbard
2026-08-04 07:01
π Exploring Retrieval-Augmented Generation (RAG) with AutoRAG! This article discusses how AutoRAG optimizes RAG for small models. It highlights the importance of tuning various settings like chunk size and retrieval methods to improve model performance. The demo showcases how AutoRAG evaluates different configurations to find the most effective way to retrieve context, ensuring better accuracy with less noise. For further insights, check out the full article! ππ #AutoRAG #MachineLearning...
Source: Red Hat Developer Blog
Isaac Tigges
2026-08-04 03:16
π¨ Red Hat Enterprise Linux 10.2 has a kernel regression causing up to 93% throughput loss on multi-NUMA instances. This issue primarily affects larger machines, while smaller instances remain largely unaffected. To mitigate this, users can disable the NI_RANDOM scheduler feature without needing a reboot: `echo NO_NI_RANDOM > /sys/kernel/debug/sched/features` For detailed insights, refer to the full article. #RedHat #Linux #KernelRegression #Performance #TechUpdate
Source: Red Hat Developer Blog
Sayali Bhavsar
2026-08-03 13:54
Deploy WordPress efficiently using Red Hat Hardened Images and image mode for RHEL. This approach minimizes vulnerabilities, allowing developers to focus on features rather than constant patching. Traditional LAMP stacks can lead to unnecessary complexity and security alerts. By utilizing Red Hat Hardened Images, you create a reproducible and maintainable WordPress environment. Learn more about building a secure WordPress stack today! π»π #WordPress #RedHat #DevOps #CyberSecurity #OpenSource
Source: Red Hat Developer Blog
Louis Imershein
2026-08-03 07:01
Explore the benefits of multitenant AI inference using dynamic resource allocation (DRA) on OpenShift! π This approach combines DRA with NVIDIA's Multi-Instance GPU (MIG) technology, allowing multiple AI workloads to operate on a single GPU without performance interference. Learn how to run two Llama 3.1 8B inference services concurrently on one NVIDIA H100 GPU, optimizing resource use and reducing costs. π‘ #AI #OpenShift #GPU #CloudComputing #ResourceOptimization
Source: Red Hat Developer Blog
Sai Ramesh Vanka
2026-07-31 13:16
Red Hat AI introduces Inference-time Scaling (ITS) to enhance model reliability in enterprise settings. By generating multiple output candidates in real-time, ITS helps select the most accurate responses without altering model weights. The open-source framework, its_hub, supports various techniques like best-of-N and self-consistency, making it adaptable for different tasks. This capability addresses reliability issues in AI agents, ensuring better tool selection and decision-making. Learn...
Source: Red Hat Developer Blog
Luke Inglis
2026-07-30 07:01
Unlock GPU efficiency with Red Hat OpenShift AI! π The latest version introduces llm-d flow control, optimizing model serving for multiple users. This feature prioritizes requests, ensuring critical workloads get served first, thus maximizing GPU use and protecting service levels. Key benefits include reduced hardware costs and consistent performance under heavy loads. Flow control is ideal for multitenant applications and safeguarding essential services from noisy-neighbor issues. Explore...
Source: Red Hat Developer Blog
Will Parker
2026-07-30 03:01
π Understanding behavioral testing for AI agents is crucial for ensuring their functionality. Traditional unit tests may pass, but they donβt catch behavioral failures. An agent can function correctly yet make incorrect decisions, such as calling the wrong tools or leaking PII. Behavioral testing focuses on observable agent actions, ensuring correct tool selection and response content. It complements agent evaluation, which assesses the quality of responses. Key components include golden...
Source: Red Hat Developer Blog
Andrew Donheiser
2026-07-30 03:01
π Modern IT organizations face the challenge of balancing automation speed with security. The article discusses "Just-in-time automated elevated access," a method combining Red Hat Ansible Automation Platform and ServiceNow ITSM. This approach ensures that elevated access is granted only during approved timeframes, reducing security risks. Key steps include validating change requests and automatically revoking access after tasks are completed. This method enhances control over privileged...
Source: Red Hat Developer Blog
Ben Forrester
2026-07-29 03:01
π Exploring the performance of storage live migration in Red Hat OpenShift Virtualization! This article assesses the functionality of migrating virtual machine disk images with minimal downtime. It highlights the evaluation of OpenShift Virtualization version 4.21. Key findings include: - Migration times scale linearly with the number of VMs. - Performance remains stable even under load, although caution is advised during high I/O workloads. For optimal results, schedule migrations during low...
Source: Red Hat Developer Blog
Abhishek Bose
2026-07-29 03:01
Meet Anthony, a voice-driven desktop orchestrator for Linux designed to enhance accessibility. π Anthony allows users to control their GNOME desktop using natural language, making it easier for those with motor disabilities or low vision to interact with their computers. It runs entirely on your machine, ensuring data privacy. π With features like continuous listening and voice feedback, Anthony simplifies tasks like managing windows and typing, transforming how users engage with technology....
Source: Red Hat Developer Blog
Jiri Prajzner
2026-07-28 07:01
π Discover how to build a distributed retrieval-augmented generation (RAG) pipeline with Ray Data on OpenShift AI! This new approach tackles the limitations of traditional document processing by enabling parallel execution of parsing, embedding, and storage stages. ππ Learn how to optimize performance, reduce idle GPU time, and efficiently manage resources using Ray Data's streaming execution. Check out the full example and get started today! #RayData #OpenShiftAI #RAGPipeline...
Source: Red Hat Developer Blog
Ana Biazetti, Bryan Keane, Saad Zaher
2026-07-28 07:01
π Discover the advancements in fine-tuning large language models (LLMs) with Orthogonal Subspace Fine-Tuning (OSFT). Traditional methods like Supervised Fine-Tuning (SFT) and Low-Rank Adaptation (LoRA) often lead to "catastrophic forgetting," where models lose existing skills while learning new tasks. OSFT addresses this by protecting a model's current capabilities during training. In tests, OSFT significantly improved performance, achieving 48.4% accuracy on diverse tasks compared to 23%...
Source: Red Hat Developer Blog
Oleg Silkin
2026-07-28 07:01
π Exciting news for OpenShift users! The compliance operator now supports custom metadata propagation on ComplianceCheckResult objects. This enhancement allows businesses to attach specific labels and annotations directly to compliance rules. Previously, compliance results lacked business context, leading to manual enrichment processes. Now, relevant metadata flows automatically with scan results, reducing operational overhead and improving automation. For more details on the new feature,...
Source: Red Hat Developer Blog
Watson Yuuma Sato, Suman Tripuraneni
2026-07-27 07:01
π Red Hat OpenShift is enhancing user experience with customizable dashboards designed for specific workflows. Out-of-the-box monitoring dashboards provide a baseline, but the new design allows users to tailor metrics to their needs. π Key challenges addressed include context switching between tools, data silos, and the need for role-specific views. The new drag-and-drop grid system empowers users to arrange metrics for better correlation. π The Red Hat build of Perses offers diverse...
Source: Red Hat Developer Blog
Foday Kargbo
2026-07-27 07:01
Integrating coding agents into workflows requires proper context for optimal results. The recent specifications AGENTS.md and Agent Skills aim to standardize this process. AGENTS.md acts as a project guide, detailing installation, tools, and documentation locations. It enhances coding agents' efficiency by providing a centralized context file. Best practices emphasize concise informationβlimit to under 150 lines to avoid performance issues. Review auto-generated contexts and use AGENTS.md as...
Source: Red Hat Developer Blog
Dejan Bosanac
2026-07-27 03:00
Unlock the potential of Logical Volume Manager (LVM) with shared storage! LVM is essential for managing storage in Linux environments, especially when using Storage Area Networks (SAN). It allows for dynamic allocation of shared storage among multiple hosts, enhancing resource management. Key concepts include Physical Volumes (PV), Logical Volumes (LV), and Volume Groups (VG). Recent updates improve LVM's performance in SAN settings, ensuring data integrity through effective locking...
Source: Red Hat Developer Blog
David Teigland
2026-07-24 03:01
Discover how PyTorch's Inductor compiler enhances performance by grouping operations into efficient Triton kernels. This method significantly reduces memory traffic and kernel overhead. Vertical fusion links dependent operations, optimizing GPU usage and increasing model speed by up to 10x. Explore pointwise fusion, which streamlines multiple operations into a single kernel, improving efficiency. ππ»π #PyTorch #MachineLearning #GPU #DataScience #AI
Source: Red Hat Developer Blog
Morrison Turnansky
2026-07-24 03:01
As Kubernetes usage grows, platform teams face challenges with telemetry costs and complexity. The common solution, "hard tenancy," leads to duplicated observability stacks, increasing operational overhead and cloud expenses. A more efficient approach uses Kiali's multi-tenancy with OpenTelemetry, allowing centralized telemetry collection while ensuring secure, cost-effective data routing. This method simplifies management and enhances developer experience without exhausting resources. π»π...
Source: Red Hat Developer Blog
Michaela Lang
2026-07-23 07:01
Organizations are facing memory limitations as they deploy advanced AI and analytics workloads. Traditional methods of increasing memory often lead to higher costs and complexity. Kove:SDM offers a solution by disaggregating memory from compute resources in OpenShift, allowing idle memory to be pooled and dynamically allocated without changing application code. Recent tests with the WASP Electronic Warfare pipeline validated this architecture, showing stable performance even under heavy...
Source: Red Hat Developer Blog
Chris Janiszewski, Ata Mufti
2026-07-23 07:01
Discover the importance of layered sandboxing for AI agents! π‘οΈ The article discusses how using both NVIDIA OpenShell and Red Hat OpenShift sandboxed containers enhances security. While OpenShell prevents data exfiltration, it can't stop kernel exploits. Conversely, OpenShift isolates kernel exploits but lacks network traffic filtering. Testing revealed that combining the two provides comprehensive protection against a wider range of threats. Explore how this dual approach can strengthen your...
Source: Red Hat Developer Blog
Eitan Geiger
2026-07-22 12:05
π Red Hat has launched Migration Toolkit for Applications 8.2, designed to modernize applications and streamline cluster migrations. This version introduces the mta-ops CLI for stateless OpenShift workloads, ensuring a safe migration process without altering source or target clusters. It follows a strict five-stage pipeline: Export β Transform β Apply β Validate β Deploy. Additionally, a built-in Hub OpenID Connect provider simplifies user management and reduces infrastructure requirements....
Source: Red Hat Developer Blog
Syed M Shaaf, Ramon Roman Nissen