Categories

Jacar categories — explore the topics A rocket whose eyes follow your cursor.
Technology

Headscale: the free alternative to Tailscale

Headscale es una reimplementación libre del plano de control de Tailscale. Con la versión 0.25 estable, es una opción sensata para mallas WireGuard privadas sin depender de la plataforma comercial. Cuento cómo, cuándo y dónde duele.

Methodologies

VEX: filtering vulnerability noise with context

Después de años acumulando SBOMs, el cuello de botella es filtrar qué CVEs afectan de verdad. VEX aparece como la pieza que convierte el ruido en señal, y en 2025 empieza a tener adopción real en pipelines de supply chain.

Methodologies

Semgrep: modern SAST in your pipeline

Semgrep has grown into one of the most pragmatic static analyzers in the ecosystem. A look at why it works where other SAST tools fail, and how to fit it into a pipeline without turning it into noise.

Technology

Nebula: Slack’s overlay VPN explained

Nebula es una VPN peer-to-peer liberada por Slack en 2019 que ha ido madurando sin mucho ruido. Explico cómo funciona, cuándo tiene sentido frente a WireGuard o Tailscale, y qué aprendí desplegándola entre varios nodos.

Technology

Deno 2.0: Node compatibility without losing identity

Deno 2.0 salió en octubre de 2024 con una apuesta clara: compatibilidad seria con npm, pnpm, package.json y node_modules, manteniendo la identidad del runtime. Medio año después, miramos qué ha supuesto para proyectos reales y dónde sigue cojeando.

Technology

Final NIST PQC standards: what to do with them now

NIST published the final post-quantum cryptography standards in August 2024. Six months on, it is time to move from headline to plan: crypto inventory, crypto-agility, a realistic timeline, and the typical mistakes of teams jumping in now.

Software Development

WASI preview 3: threads and async in WebAssembly

WASI 0.3, also known as preview 3, was ratified on June 11, 2026, adding native asynchronous concurrency to the WebAssembly component model through streams, futures, and async functions. It fixes old fragmentation across languages and runtimes, enables real composition between Wasm services, and paves the way for cooperative threads planned in upcoming 0.3.x releases.

Architecture

Cloudflare Workers in 2025: from edge to enterprise

Cloudflare Workers turned eight in 2025 without slowing down: it now ships D1 for databases, R2 for egress-free storage, Durable Objects for distributed state, and Workers AI for running models without managing GPUs. It remains the fastest option for edge logic; for large in-memory processes or strict global consistency, other platforms fit better.

Artificial Intelligence

NPU in the PC: faster, cheaper local AI

Qualcomm, Intel and AMD Copilot+ processors have normalised the presence of an NPU in everyday PCs. A 40 TOPS NPU can run quantised Phi-3 Mini drawing just 5-10 W, versus 40-50 W for a laptop GPU doing the same task. What actually changes for running AI models locally, and when it is worth it.