Site map Everything published on jacar.es on one page: the learning paths, the hands-on labs and every article, grouped by category. Pages AboutCookie PolicyLabLocal LLM CalculatorPrivacy Policy Learning paths Advanced Neural Network OptimizationAgent Ecosystem: MCP, Gateways and PlatformsAI Agents FundamentalsAI Agents in Production: Evaluation and ReliabilityAI Coding AgentsFrameworks to Build AI AgentsLocal LLMs: run models on your own hardwareLoss Functions in Neural NetworksMathematical Foundations of Neural NetworksMonitoring and observability with DockerNetworking and secure remote access with DockerSelf-hosted Agentic Models and Tool CallingSelf-hosted dashboards and notifications with DockerSelf-hosted databases and storage with DockerSelf-hosted docs and productivity with DockerSelf-hosting with Docker: from zero to productionStability, Initialization and RegularizationThe Neuron and Activation FunctionsTraining: Gradient Descent and Backpropagation Hands-on labs How to build an agentic VM with herdr and several AI CLIsHow to Run a Private, Self-Hosted Docker RegistryMulti-Stage Docker Builds: Smaller, More Secure ImagesDocker Healthchecks and Restart PoliciesManaging Secrets in Docker (without baking them into the image)Traefik Reverse Proxy with DockerPihole in Docker (DNS Sinkhole)Docker Volume Backup and RestoreHow to Build a Multi-Container App with Docker ComposeHow Data Persists in Docker with Volumes and Bind MountsDocker Networking (Bridge and Custom)Install Portainer with Docker Compose v2 Articles by category AI Agents (8) How to migrate your MCP server to FastMCP 4 and the Python SDK 2How to use Shieldstral as a local guardrail for your agentHow to use OpenCode with local models on llama.cpp and OllamaHow to migrate self-hosted Langfuse from v3 to v4How to use DeepSeek Harness with a local modelHow to build an agent with Dify Agent and a local modelEngram: persistent memory for coding agentsHerdr: a terminal multiplexer for coding agents Architecture (91) Kubernetes 1.35 GA: an operations-side balance sheetSkills and subagents: the agent reuse patternMCP as multi-vendor standard: patterns already matureHybrid RAG in 2026: the patterns that keep winningDocker Swarm in 2023: When It Still Makes SenseAgent OS in production: real cases without the marketingEnterprise GraphRAG: patterns after a year of adoptionConsolidated MCP ecosystem: a quick map for 2026Consolidated platform engineering: who wins and who gets stuckAgent-to-agent protocols v1: what we have in handKubernetes 1.35: what you can already see comingcontainerd with Wasm: mixed workloads in productionAgent-to-agent protocols: the next open layergVisor: sandboxing for multi-tenant containersLLM caches: saving tokens without dropping qualityInference routers: choosing a model based on the requestTigerBeetle: a database built for financial transactionsPlatform engineering: consolidation after the boomCitus: scaling Postgres horizontally without leaving itSQLite in production: patterns that have aged wellModern data engineering: dbt, Iceberg and the lakehouse come togetherDuckDB in enterprise analytics: concrete casesAgent OS: the concept shaping the new stack layerWolfi: the base distribution designed for containersKata Containers: strong isolation with a Docker-like experienceModel Context Protocol in 2025: from announcement to ecosystemYugabyteDB and CockroachDB: distributed databases in 2025Redpanda: an alternative to Kafka with a different architectureKubernetes 1.34: a summary for teams with little timeRedis 8.2 and its vector support: when it actually makes senseKafka without ZooKeeper: KRaft in productionDelta Lake and Apache Iceberg: 2025 comparisoncontainerd 2.0 in production: real migrationsFirecracker: microVMs for multi-tenant servicesKubernetes 1.32: what the first jump of 2025 bringsPostgreSQL 17: optimisations that change real queriesValkey as a Redis replacement: a real migration with Valkey 8.1Kubernetes 1.33: the sneak peek from an operations lensApplying graph RAG to a real productCoolify: a self-hosted Vercel on your own infrastructureMicrosoft’s GraphRAG in enterprise: patterns that workCloudflare Workers in 2025: from edge to enterpriseHow to install JuiceFS as a shared filesystemHybrid Search: Combining BM25 and Vectors SeriouslyModel Context Protocol: Anthropic’s Open ProposalMariaDB 11.7: The Fork That Keeps Its Own PathDuckDB: Fast Analytics Without Moving DataKubernetes 1.30: The Improvements Operators Actually AppreciatevLLM: Serving LLMs in Production with Very High ThroughputRAG in Production: Patterns That Work and Those That Don’tKubecost and OpenCost: Native FinOps in KubernetesLitestream: Near-Real-Time Replication for SQLiteCloudflare Workers in 2024: KV, D1, and the New Edge StackKubernetes 1.31: the stabilisations that matter day to dayOpenTelemetry: Stable Logs for Complete TelemetryService Mesh in 2024: Istio Ambient and Cilium MeshValkey: The Open Fork After Redis’s License ChangeContainer Monitoring: Beyond cAdvisorPostgreSQL 17: The Novelties That Show PromisePostgreSQL 16: Logical Replication That’s Now PracticalModern SCADA in Containers: Advantages and RisksCilium Service Mesh: When You Don’t Need SidecarsSQLite in Production: Not Just for MobileLinkerd: The Pragmatic Service Mesh AlternativeHow to Install PostgreSQL with pgvector Step by StepBackstage, Port and Cortex: Three Paths to the IDPpgvector in 2024: HNSW Indexes and Real ScalingSQLite and DuckDB: When Each Is the Right ChoiceGitOps With ArgoCD: From Hype to Stable ProductionService Mesh in 2023: Istio, Linkerd, and the Cilium Optioncontainerd: The Runtime Underpinning KubernetesEvent-Driven Architecture: When and How to Adopt ItBackstage: Spotify’s Developer PortaleBPF: Kernel Observability Without RecompilingVector Databases: Qdrant, Pinecone, and WeaviatePostgreSQL 16: Changes That Affect Day-to-Day Workpgvector: Semantic Search Without Leaving PostgresChroma: A Lightweight Vector Database for Embedding PrototypesMicro-frontends in Practice: Benefits and PitfallsPlatform Engineering: Internal Developer PlatformsRedis: Caching Strategies Every Backend Should KnowKubernetes 1.27: The Changes That Matter to OperatorsKafka in 2023: Event Streaming in the EnterpriseOpenTelemetry: Unifying Logs, Metrics, and TracesHashiCorp Vault for Secrets ManagementCilium and the Future of Container Networking with eBPFRabbitMQ for Message Queues: When It’s Still the ChoiceKubernetes 1.28: Sidecar Containers as First-Class CitizensPixie: Native Kubernetes Observability Powered by eBPFFrom Monolith to Microservices: Transforming the ArchitectureModules vs. Microservices: The Architecture Battle Artificial Intelligence (221) How to red-team a local LLM with garak 0.17 and OllamaHow to fix “typical_p is no longer supported” in OllamaMaple-Preview, MiniCPM5-2B or Spark-X2.5, which CPU-only LLM to pick for an arm64 machineHow to import a Hugging Face model into Ollama 0.34What Colibri is and how it runs giant MoE models by streaming experts from diskThinkingBox: checking whether your agent does real workWhy many code agents use agentic search instead of a vector databaseAgent Lightning: reinforcement learning for agents without rewriting themWhat the new MCP specification breaks in your serverLocal multimodal models: the VRAM arithmetic on a 16 GB cardThe Apple Watch Ultra 3 as an AI assistant in EuropeGemma 4 locally, which size fits your GPUWhich open models of August 2026 are worth running on your own hardwareMuse Code: Meta’s terminal coding agentEverything that changed in Claude Code over the summer of 2026What is oMLX and how it differs from MLX, Ollama and LM StudioWhat FreeToken is and how it runs 753B MoE models on one machineOpenRouter: A Gateway for AI Modelsbrowser-use: agents that browse the webFirecrawl: Web Data for AgentsComposio: Tools and Integrations for AgentsDify: a self-hosted LLMOps platformFlowise: self-hosted low-code agentsLangflow: a visual agent builderBuilding Your Own MCP ServerDurable Agent Execution with TemporalDeploying an AI Agent to ProductionE2B: a code sandbox for agentsPrompt caching to cut costspromptfoo: testing prompts and agentsDeepEval: evaluating AI agentsHelicone: LLM observability in one lineLangfuse: self-hosted agent observabilityDSPy: optimizing prompts and LLM programsConstrained generation with OutlinesInstructor: reliable structured outputsOpen models with tool calling: which to chooseFunction calling with Ollama on your own machineQwen-Agent: tool use with Qwen modelsNous Hermes 4: The Agentic Model with Hybrid ReasoningNous Hermes 3: an open-weight agentic modelHaystack: Pipelines and AgentsAgno: a high-performance agent frameworkMastra: Agents in TypeScriptMicrosoft’s Semantic KernelBuilding Agents with LlamaIndexGoogle ADK: The Agent Development Kit for Building AgentsHugging Face smolagents: Agents That Think in CodePydantic AI: typed agentsThe OpenAI Agents SDKContinue: The Open-Source Coding AssistantWindsurf: The Agentic EditorGemini CLI: An Agent in Your TerminalGoose: Block’s Coding AgentOpenHands: An Autonomous Coding AgentOpenAI’s Codex CLIHow to Use Roo CodeHow to Use Cline in VS CodeMulti-Agent System PatternsPlan-and-Execute versus ReActContext Engineering for AgentsHuman-in-the-Loop in AI AgentsMemory in AI Agents: Short and Long TermThe Reflection Pattern in AI AgentsPlanning and Task Decomposition in AgentsThe Agentic Loop and the ReAct PatternWhat Is an AI Agent?What Docker Agent is, the CLI plugin for building AI agents in YAMLWhat Open GSD is, the Git-Ship-Done loop for coding agentsWhat is a vector embedding and what is it used forRAG with Postgres and pgvector in production: from PoC to SLOEU AI Act 2026: a technical checklist for Spanish CTOsAgent observability with OpenTelemetry GenAI semconv in 2026How to install and tune oMLX on M5 Max 128 GBMulti-agent systems: LangGraph vs CrewAI vs Autogen in 2026How to build a production-ready agent with the Anthropic SDK, step by stepClaude Code vs Cursor vs GitHub Copilot in 2026: a comparison with measured tasksMCP (Model Context Protocol) in 2026: the complete guide for engineering teamsFinOps on agent tokens: the invoice that surprisesProfitable niche AI startups: the patterns that repeatRuntime-generated UI: the first serious yearDPO and alternatives to RLHF: practical state in 2026Synthetic training data in 2026: when it worksMature LLM-as-judge: when to trust and when notClaude Sonnet 4.6 in production: the cost-quality balanceClaude Opus 4.7 and long-horizon tasks: real changesAI agent incidents: recovery runbooks that workLLM red teaming: a practical playbookProduction-grade agent evaluations: the framework that worksPrompt Engineering: From Trick to Mature DisciplineDeploy Llama 3.3 and Mistral locally with Ollama and Open WebUI on Ubuntu 24.04How to install a local MCP server for your editorEnterprise agent governance: the controls that are no longer optionalLessons from agents in production in 2025: summary for 2026European AI Act: full application and lessons from the first cycleFinOps for AI workloads in 2026: the real painAgents that drive the computer: patterns that workAI incident postmortems: what they have taught usAI startup market correction in 2026Claude Haiku 4.5: lightweight power for massive agent fleetsKnowledge graph renaissance with LLMsUX for agents: first design consensusSovereign AI in Europe: practical statusPhi-3 on the edge: Microsoft’s SLM in 2025LLM-generated documentation: when it helps and when it gets in the wayLLM guardrails: frameworks and their real costAI agent observability: tools and what to instrument firstTesting with AI: the determinism problemGPT-5: public availability and early impressionsEuropean AI Act: enforcement kicking in from August 2025SLMs at the industrial edge: when the small model is betterRAG 2.0: knowledge graphs, vectors, and hybridComputer Use in production: agents that drive the interfaceMCP clients in editors: AI integrates where you already workContinuous integration with AI agents: early patternsGemini 2.5: context scaling and multimodalityThe initial Claude 4 family: first quality testsUI design for agents: principles we’re starting to understandCommunity MCP servers: which ones are worth itPrompt injection defense: what actually worksThe knowledge graph era is reborn with LLMsFinOps applied to AI: where the cost really goesContinuous evaluation of RAG: dashboards that actually matterStartup funding 2025: everything labeled AI is going upLLM wrappers: when they are a business and when they are notLLM agent security: the new class of threatsAI agents in the enterprise: from demo to measurable valueFigma AI: how product design is changingAI governance in enterprise: committees, policies, auditsClaude 3.7 Sonnet: the intermediate step toward the 4 familyOpen-weight models in enterprise: one year onvLLM in 2025: the improvements that matter to LLM-serving teamsAlignment evaluation: RLHF, DPO, and recent alternativesGemma 2: Google’s open model one year latero3 in public: the reasoning leap is confirmedGemini 2.0: integrated tools and agent modeAI-assisted code review: an honest adoption storyLlama 3.2 at the edge: Meta bets on smallNPU in the PC: faster, cheaper local AIHow to Evaluate a RAG System Without Fooling Yourselfo1-preview: OpenAI’s Model That Thinks Before Answeringllama.cpp: Optimisations That Keep SurprisingOllama in 2024: Running LLMs Locally Without PainProduct-Market Fit in the AI Era: What ChangesLLM Observability: Traces, Costs, and QualityTensorRT-LLM: Extreme Acceleration on NVIDIA GPUs for LLMsLoRA and QLoRA: Efficient Fine-Tuning on a Single LaptopClaude’s Computer Use: When the Agent Moves the MouseGitHub Copilot Workspace: GitHub’s Conversational IDESwarm: OpenAI’s Experiment for Multi-Role AgentsClaude 3.5 Sonnet: The Model That Rewrote Price-Quality BalanceMistral Large: European Contender Against GPT-4CrewAI: Orchestrating AI Agent TeamsOpenAI Assistants API: Stateful Agents Without Your Own InfrastructureEU AI Act: What Changes for Your CompanyHow to Install Ollama on macOS with Apple SiliconGitLab Duo: AI Assistants Integrated in Your PipelineLlama 3.1 405B: When Open Catches Up With Closed Top-TierRe-Ranking in RAG: The Piece That Really Raises QualityGPT-4 Turbo: Long Context and More Reasonable CostsRetrieval Evaluation Frameworks: Ragas and SimilarSGLang: Fine Control Over LLM ExecutionGPT-4o: OpenAI’s Native MultimodalityLlama 3: Meta’s New Open Standardnomic-embed-text: Competitive Open EmbeddingsLangGraph: State Graphs for More Robust AgentsConstrained Decoding for Structured LLM OutputsClaude 3 Family: Haiku, Sonnet and Opus ComparedMixtral 8x22B: Open and Powerful Mixture of ExpertsLM Studio: Exploring AI Models from Your DesktopONNX Runtime at the Edge: Portable, Fast InferenceLiteLLM: A Proxy to Unify Model ProvidersGemini 1.5: Millions of Tokens of Context in ProductionChoosing an Open LLM for Enterprise in 2024OpenAI text-embedding-3: What Changes vs the Previous OneCohere Embed v3: Multilingual and Enterprise-OrientedHugging Face TGI: Serving Open Models at ScaleClaude 2: Anthropic’s Alternative to GPT-4Model Quantization and llama.cpp on Your LaptopLangChain: The Framework for Orchestrating LLM ApplicationsText Embeddings: Turning Words Into Useful VectorsOpenAI Function Calling: Structuring Model OutputMidjourney v5: Photorealistic Quality at Prompt’s ReachGenerative AI and Regulation: First Legislative StepsHow to Install Ollama to Run LLMs on Your ComputerPredictive Maintenance with Classic Machine LearningGPT-4 In Depth: Real Capabilities vs ExpectationsLLaMA 2 and the New Wave of Open Language ModelsBard and PaLM 2: Google’s Bet on Generative AILLM Fine-Tuning: When It’s Worth Training Your OwnStable Diffusion XL: Powerful Open Image GenerationChatGPT With Plugins: An Ecosystem Under ConstructionOpenAI Code Interpreter: Conversational Data AnalysisDINOv2: Advances in Self-Supervised Computer VisionCerebras-GPT: 7 Open-Source LLM Models Ready to UseVector Database Comparison: Qdrant, Pinecone, and WeaviateEnsemble Learning in Machine LearningThe Hyperbolic Tangent: A Powerful Activation FunctionThe Sigmoid Function: A Key Tool in Neural NetworksSoftmax Function: Activation for ClassificationThe Leaky ReLU Function and Its Role in Neural NetworksThe Rectified Linear Unit (ReLU): An Essential Tool for Deep LearningThe Step Function: An Essential Tool in Neural NetworksLinear Function: A Common Activation FunctionThe Fully Connected Neural Network: An Innovative Approach to Machine LearningMathematical Formulation of Artificial Neural Network InputMultilayer Neural Networks: Advancing Artificial IntelligenceDataFrames and Pipelines in Spark: Data Processing OptimisationLazyPredict in Python Example: Automating the Machine Learning ModelB2B Sales Optimisation with AIChatGPT 4: Advances in Chatbot TechnologyPre-trained Models and Transfer LearningAdversarial Machine Learning: Challenges and SolutionsFederated Learning and Privacy: Data ProtectionRecommendation and Collaborative Filtering: Effective SystemsExplaining AI Through XAIReinforcement Learning: An Autonomous Learning TechniqueRobotics and Intelligent Automation: The New Industrial EraImage Analysis: Computer VisionNLP Advances: The Technology Revolutionising Language ProcessingNeural Networks and Deep Learning: Advances in Artificial Intelligence How to Install (29) How to install OpenBao with Docker and Raft storageHow to enable Paperless-ngx 3 local AI with OllamaHow to install OpenClaw with Docker and a local modelHow to upgrade Jellyfin 12 with Docker and keep a way backHow to migrate from MinIO to Garage with DockerHow to install llama.cpp on Linux, macOS and DockerHow to install Forgejo with Docker, Caddy and Forgejo RunnerHow to install Jellyfin with DockerHow to install k3s in your homelabHow to install Karakeep, your bookmark manager with AI taggingHow to install Pangolin and expose your services without Cloudflare TunnelHow to install Immich with Docker and leave Google Photos behindHow to install Pocket ID and get passkey SSO on your homelabHow to install Komodo with Docker and manage several hostsWhat PegaProx adds over the Proxmox VE 9 GUIHow to install Coolify on Docker (2026 step-by-step guide)How to install Portainer with Docker Compose v2How to install Uptime Kuma for basic monitoringHow to install restic for encrypted backupsHow to install n8n self-hosted with DockerHow to Install Authentik for Self-Hosted SSOGenerics in Go: three years later, what has survivedHow to Install CrowdSec as a Community WAFHow to Install Traefik on Docker Swarm with CertificatesHow to Install Docker on Debian 13 Step by StepHow to Install Docker on Ubuntu 24.04How to Install Traefik with Docker ComposeHow to Install Docker Compose on Ubuntu 20.04How to Install Docker on Ubuntu 20.04 Industry 4.0 (22) Collaborative robotics on the factory floor: 2026 balance sheetHumanoid robotics: beyond the viral videosIndustry 4.0 and data sovereignty: the 2026 tensionAdditive manufacturing: industrial state at the start of 2026Dark factory: robotics and automation todayPrivate 5G networks: enterprise deployments in 2025Digital twins on the shop floor: the 2025 harvestMetal additive manufacturing: where it stands in 2025IEC 62443: OT Cybersecurity Explained for IT TeamsIndustrial Edge Computing: Low Latency Where Data HappensPLC and OPC UA: Modernising the Plant Without Stopping ProductionDigital Twins in Health: Beyond MarketingSustainable Data Centers: What Changes in 2024Industrial ‘As a Service’: Models That WorkPrivate 5G Networks in Factories: When It Makes SenseDigital Twins in Energy: Operating Grids Without SurprisesDigital Thread: Industry 4.0 Beyond the TwinDigital Twins: When the Factory Has a Software ReplicaIndustry 4.0: IoT Sensors as the Operational Data FoundationThe Customer Digital Twin: A Strategic ToolThe 10 Best Digital Twin ToolsIndustry 4.0 Concepts: The Digital Twin of the Organization Mac (9) Benchmarks and the memory ceiling in oMLXThe oMLX admin dashboard and its command lineThe oMLX API key, port and endpointsInstalling, updating and uninstalling oMLX with HomebrewModel management in oMLX: downloads, aliases, TTL and disk cacheoMLX as an MCP server: control your local models from ClaudeEssential Software for Your New M5 Mac (2026 guide)Apple M4 Pro on developer machines: real-world experienceApple Silicon M3 and M4: The Silent Advance in Portable Computing Methodologies (39) AI-integrated DevOps tools in my daily flowRICE: a prioritization framework for product roadmapsPost-quantum migration: what is actually happeningProduct discovery with AI: practices that stickCarbon-aware scheduling by default: first balanceSRE with AI: dashboards that actually helpCarbon-aware computing: now the default behaviorUser research in the age of generative AIMigrating SSH to post-quantum cryptography: a practical guideContinuous profiling with eBPF in productionThe Site Reliability Workbook: patterns we still useVEX: filtering vulnerability noise with contextSemgrep: modern SAST in your pipelineZero Trust integrated with SIEM: what actually worksDependabot and Renovate: two approaches to updating dependenciesPair programming with AI in 2025: habits that stickSLSA v1.0: a mature framework for the software supply chainGreen Software Principles: A Checklist for TeamsAlertmanager: Routing That Doesn’t Wake Your Team at 3amChaos Engineering in Enterprise: Beyond Chaos for Chaos’s SakeAnsible and Pulumi: Two Automation Philosophies CoexistingCarbon-Aware Computing: Reduce Emissions Without Rebuilding EverythingSigstore in Image Registries: Adoption and RealityObservability and SLOs: Error Budgets That Get MetBlameless Post-Mortems: How to Actually ImproveSaaS Consolidation: When Lock-In Becomes RiskPractical DevSecOps with Sigstore and cosignFlux CD vs ArgoCD: Which to Choose for Your PlatformSLSA Level 3: Hardening the Software Supply ChainFinOps: Controlling Cloud Cost Without Slowing the TeamApplying Google’s SRE Book Without Being GoogleContract Testing with Pact for MicroservicesPrometheus: Writing Alerts That Won’t Get IgnoredMoSCoW Method: Effective Requirements PrioritisationThe Kano Model: Improving Customer SatisfactionDesign Thinking: Methods and StrategiesOKR Methodology: Maximise Your ObjectivesAgile Methodologies: Optimising Project DevelopmentMethodologies for Defining Objectives: A Strategic Approach Software Development (36) AI editors in 2026: comparison after a year of useAI tools for developers: the 2026 stackRust in the Linux kernel: balance after several yearsWASI preview 3: adoption and real casesNPUs for developers: what you can actually do todayAstro 5: when content and applications convergePython 3.13 with optional GIL: what it means for teamsQwik in production: resumable and cheap on the clientTypeScript 5.5: advanced types without breaking thingsHTMX in the enterprise: where it fits and where it does notPolars versus pandas in 2025: the real-world practiceRust in Linux: stable drivers and first serious contributionsRust Edition 2024: what really changes day to dayFull-stack TypeScript in 2025: the good, the meh, the badWASI preview 3: threads and async in WebAssemblyPython 3.12: Incremental Acceleration and Errors That Finally HelpRust 1.75 and 1.76: Improvements Noticeable DailyQwik: Betting on Resumability Instead of HydrationSvelteKit 1.0 a Year Later: Real Adoption and LimitsAider: AI for Refactoring from the TerminalHTMX: Interactivity Without SPA FrameworkAstro and Islands: The Web That Sends Less JavaScriptRust in Linux: Experimental Drivers Opening the WayWASI 0.2 GA: Truly Composable WebAssemblyRemix v2: The React Current Centered on Web StandardsNext.js App Router: Lessons from Migrating Real ProjectsGo 1.22: Updates That Simplify Idiomatic CodeTypeScript 5.4: More Powerful Types, Fewer TricksZed: A Modern Editor Built for CollaborationCursor: The Editor Built Around AICorporate Low-Code: Retool and Appsmith for Internal ToolsWASI 0.2: The New Standard Interface for WebAssemblyRust in the Linux Kernel: First Steps and ControversiesRust for Backend: axum and tokio in Real ServicesWebAssembly: The Component Model as the Next FrontierWord Choice in Your App: Key to the User Experience Startup (7) NIS2 in Spain: a technical translation of 2026 obligationsFounders outside the Silicon Valley ring: 2026 alternativesStartup Funding in 2023: Reality After the CorrectionY Combinator 2025: trends from the AI cohortsY Combinator Demo Day 2024: Trends Left by UnicornsStartup Funding in 2024: The Uneven RecoveryUseful Metrics for Startups: Types and Uses Technology (116) CVE-2026-80521 container escape: check your Docker and k3s hosts and mitigate itDropout and Its Mathematical InterpretationLayer Normalization and VariantsBatch NormalizationWeight Initialization: Xavier/Glorot and HeSecond-Order Methods: Newton and Its ApproximationsLearning Rate Schedules and WarmupAdam and AdamW: The Default OptimizerAdaGrad and RMSProp: Adaptive Learning RatesNesterov Accelerated Gradient (NAG)Momentum in Gradient DescentLocal Minima, Saddle Points and the Loss LandscapeGradients of a Dense Layer: Matrices and JacobiansThe Computational Graph and Automatic DifferentiationBackpropagation: The Step-by-Step DerivationBackpropagation: The IntuitionBatch, Stochastic and Mini-Batch Gradient DescentThe Learning Rate in TrainingGradient DescentL1 and L2 Regularization (Weight Decay)Entropy, Cross-Entropy and KL DivergenceCategorical Cross-EntropyBinary Cross-EntropyMean Absolute Error (MAE) and Huber LossMean Squared Error (MSE) as a Loss FunctionWhat Is a Loss Function and a Cost FunctionForward Propagation in a Multilayer NetworkThe Exploding Gradient ProblemThe Vanishing Gradient ProblemHow to Choose an Activation Function (Comparison)The Mish Activation FunctionThe Softplus Activation FunctionThe SELU Activation and Self-Normalizing NetworksThe ELU (Exponential Linear Unit) Activation FunctionThe Swish (SiLU) Activation FunctionThe GELU Activation Function in Neural NetworksWhat Is an Activation Function and Why It Is NeededWeights, Biases and a Neuron’s Weighted SumThe Perceptron, the Artificial Neuron and Its MathsProbability Essentials for Neural NetworksThe Exponential and Natural Logarithm in Deep LearningPartial Derivatives and the Gradient in Neural NetworksThe Chain Rule, the Engine of BackpropagationEssential Differentiation Rules for Neural NetworksDerivatives, the Rate of Change That Teaches the NetworkVector Norms (L1, L2) and DistanceTranspose, Identity and Inverse MatricesMatrix Multiplication in Neural NetworksThe Dot Product and the NeuronScalars, Vectors, Matrices and Tensors ExplainedWhat Mathematics Is Behind Neural NetworksNVIDIA alternatives in 2026: where the market is headingNext-generation NPUs: the hardware moving AI in 2026XR, AR and VR in 2026: the honest state after the cycleBlockchain out of the hype: the uses that remain in 2026Observability tools I would recommend in 2026Quantum computing: real progress without hypeFly.io: deploying globally without complicating your lifeMicrosoft Garnet: a high-performance cache alternativeOSV-Scanner: vulnerabilities with a source of truthCVE-Based Attack Surface Management: 2025 MaturityUnikraft and unikernels: the promise that returnsPost-quantum cryptography in TLS: real-world adoptionDragonfly: the modern cache inspired by RedisNVIDIA Blackwell GPUs: what changes for trainingParca, Beyla and Grafana: a sidecar-free observability stackMesh networks with WireGuard without losing your mindHeadscale: the free alternative to TailscaleNIS2: what we learned from the first year of applying itCodeQL in GitHub Advanced Security: what it actually coversNebula: Slack’s overlay VPN explainedDeno 2.0: Node compatibility without losing identityFinal NIST PQC standards: what to do with them nowVector: A Log Agent Worth TryingTerraform 1.6 Onward: Updates After the License ChangeeBPF for Continuous Profiling: Parca and BeylaNIST PQC: The Post-Quantum Cryptography StandardsTrivy and Grype a Year Later: Which Matured BetterDocker Scout: Vulnerabilities from Build to RegistryGrafana Beyla: Auto-Instrumentation Without Touching CodeSnapdragon X Elite: ARM Arrives at Productivity PCsOpenTofu in Production: The First Year After the ForkNIS2: Transposition Status Across Member StatesParca: Open eBPF-Based Continuous ProfilingChainguard Images: Minimal and Signed ImagesDeno Deploy: TypeScript at the Edge Without a ServerLoki at Scale: Lessons from High-Volume LogsFastly Compute: High-Performance Edge with WebAssemblyFalco: Runtime Threat Detection with eBPFCloudflare Workers: Edge Compute Without ContainersZero Trust: Principles to Stop Trusting the NetworkTerraform’s License Change and Its ImplicationsOpenTofu: The Open Response to Terraform’s License ChangeTrivy and Grype: Container Image Scanning in CISupply-Chain Attacks: Lessons from 2023The Grafana Stack: Loki, Tempo, and Mimir for Open Observabilitynerdctl: A Lightweight Docker Alternative Over containerdNIS2: What Europe’s New Directive Changes for CybersecurityWeb Accessibility: Covering WCAG Without Losing Your MindeBPF: High-Performance Monitoring in LinuxIs the CIO Role Outdated? The Evolution of the Chief Information OfficerThe 10 Best RPA Tools: Robotic Process Automation SoftwareTrends: Financial TechnologyThe 3D Printing Revolution: Transformative TechnologyExploring Augmented Reality: Innovative TechnologyThe Power of Big Data in Decision-MakingRenewable Energy: Advances and DevelopmentsIoT: The Intelligent Connection of the WorldCybersecurity: Protection Against Digital ThreatsDevelopment and Advances in Artificial IntelligenceBlockchain Technology: Fundamentals and ApplicationsCryptocurrencies: Technical and Professional AnalysisProgramming with GitHub CodespacesMicrosoft 365 Copilot: The Technical Assistance ToolGitHub Copilot: The Future of AI ProgrammingIndustry 4.0: The Digital Industrial Revolution Tools (82) How to upgrade Gitea 1.27 to Gitea 28 with Docker without breaking webhooks or mirrorsHow to install Gatus with Docker Compose, monitoring as code vs Uptime KumaHow to fix pull access denied for minio/minio in DockerHow to install Open WebUI with Docker and Ollama, slim image includedHow to migrate from MinIO to RustFS 1.0 with DockerRabbitMQ 4.3 streams vs quorum queues, measuredHow to use init containers in Docker Compose with pre_startFile Browser alternatives now that the project is archivedHow to enable CrowdSec 1.8 bot detectionWhy Docker Compose 5.5 recreates your containersHow to prepare your self-hosted n8n for n8n 3.0What the TrueNAS Proxmox plugin does and when it pays offActivepieces and Windmill against n8n: which self-hosted automation to choosek3s vs k0s vs Talos Linux: which lightweight Kubernetes to chooseWhat Obscura is, the Rust headless browser for AI agentsClaude Code vs Codex CLI vs Muse Code in 2026Pocket ID vs Authelia vs Authentik: which SSO to pick for your homelabKomodo vs Portainer vs Dockge: which Docker manager to choose in 2026Portainer Agent: managing a second Docker host from one consoleHow to Install SearXNG with DockerHow to Run Headscale with DockerHow to Install Home Assistant with DockerHow to Install Gitea with DockerHow to Install Vaultwarden with DockerHow to Set Up a WireGuard VPN with wg-easy in DockerHow to Install Nginx Proxy Manager with DockerHow to Install AdGuard Home with DockerHow to Install Pi-hole with DockerHow to Monitor Website Changes with changedetection.io in DockerHow to Install IT-Tools with DockerHow to Install Gotify with DockerHow to Install ntfy with DockerHow to Install Dashy with DockerHow to Install Heimdall with DockerHow to Install Homarr with DockerHow to Install Homepage Dashboard with DockerHow to Install Beszel with DockerHow to Monitor Your Server with Glances in DockerHow to Install Netdata with DockerHow to View Container Logs with DozzleHow to Centralize Logs with Grafana Loki in DockerHow to Install Grafana with DockerHost and Container Metrics with Node Exporter and cAdvisorHow to Install Prometheus with DockerHow to Install Duplicati with DockerHow to Install Seafile with DockerHow to Install Syncthing with DockerHow to Install MinIO with DockerHow to Manage Databases with Adminer and pgweb in DockerHow to Install Redis with DockerHow to Install MariaDB with DockerHow to Install PostgreSQL with DockerHow to Install Firefly III with DockerHow to Install Stirling-PDF with DockerHow to Install Vikunja with DockerHow to Install Outline with DockerHow to Install Docmost with DockerHow to Install Wiki.js with DockerHow to Install BookStack with DockerHow to Install Paperless-ngx with DockerHow to Install Nextcloud with DockerWatchtower: Automatically Update Docker ContainersHow to Run a Private Docker Image RegistryResource Limits (CPU and RAM) and Logging in DockerProfiles in Docker ComposeHealthchecks and Restart Policies in Docker ComposeEnvironment Variables and Secrets in Docker ComposeDocker Volumes and Bind Mounts: Persisting DataDocker networking: bridge, host and custom networksDokku: the small PaaS that never diesRailway and Render: deployment platforms without surprisesSpotify’s Backstage fork: what changes for usersDokploy: lightweight deployment on top of Docker SwarmHome lab: self-hosted lab as a testing groundFigma Dev Mode: From Design to Code with Less FrictionFluent Bit: Lightweight Log Collection in ProductionSemaphore: Ansible UI When the Team GrowsPodman: Containers Without a Daemon or RootFigma: Collaborative Prototyping for Product TeamsEfficient PC Management with Microsoft PC ManagerGitHub Copilot: The Intelligent Code AssistantRustDesk Remote Desktop Tool: Professional Solution User Experience (4) European Accessibility Act: the first year in practiceAccessibility overlays: the 2025 criticisms explainedWCAG 2.2: What the New Accessibility Version BringsEuropean Accessibility Act: Preparing for June 2025
