Category: tech
184 insights in this category - page 4 of 10. View all insights
Parlor: On-Device Real-Time Voice and Vision AI
Parlor runs real-time voice and vision AI conversations locally using Gemma 4 E2B and Kokoro TTS, with usable latency on an Apple M3 Pro and zero server costs.
OpenScreen: Free Open-Source Alternative to Screen Studio
OpenScreen is a free, open-source Screen Studio alternative for product demos and walkthroughs. Core recording and editing, no subscriptions, no watermarks.
Hermes Agent - AI Agent Framework
Hermes Agent by NousResearch is a self-hosted autonomous agent with persistent memory, multi-channel messaging, and an open skill ecosystem that compounds.
Why Switzerland Has 25 Gbit Internet and America Doesn't
Switzerland gets 25 Gbit/s residential fiber by mandating shared neutral infrastructure open to any ISP, while the US and Germany produce monopolies.
Harness Engineering: Beyond Prompt and Context Engineering
Harness engineering is the scaffolding layer that lets an AI agent run for hours or days, decompose work, self-prompt at each step, and evaluate quality.
How We Built a Virtual Filesystem for Our Assistant
Mintlify replaced sandbox containers with ChromaFs, a virtual filesystem that maps UNIX shell commands to Chroma vector queries, cutting session creation 460x.
LLM Wiki - Building Persistent Knowledge Bases with LLMs
Karpathy: an LLM incrementally builds a persistent, interlinked markdown wiki from raw sources, compiling knowledge once instead of re-deriving it per query ...
AI state of the union: we passed the inflection point
Simon Willison on Lenny's Podcast: AI coding agents crossed an inflection point in November 2025 with Claude Opus 4.5 and GPT 5.1. Dark factories are coming.
Sol: Free Open-Source macOS Launcher
Sol is a free, open-source macOS launcher consolidating app search, clipboard history, window management, calendar integration, and more into one keyboard-dr...
Email address obfuscation: What works in 2026?
A 2026 benchmark of 15+ email obfuscation techniques, using a live honeypot to measure which methods actually stop spam harvesters, with code examples.
Gemma 4: Google's Open-Weights Model for Mobile and IoT
Google DeepMind's Gemma 4 targets mobile and IoT deployment with multimodal input, native function calling for agents, and fine-tuning support.
Gemma 4 Has Landed
Google released Gemma 4 as four Apache 2.0 models with native vision, function calling, reasoning, and audio on edge, closing the open-weights gap.
Google DeepMind Gemma 4 - Open-Weights Models for On-Device AI
Google DeepMind's Gemma 4 is an open-weights family for on-device and edge deployment with multimodal input, native function calling, and multilingual context.
EmDash: WordPress successor that solves plugin security
Cloudflare released EmDash v0.1.0, a TypeScript/serverless CMS isolating plugins in sandboxed V8 workers with capability-based permissions and x402 payments.
LinkedIn Is Illegally Searching Your Computer
LinkedIn runs hidden JavaScript that scans visitors' browsers for extensions and software, transmitting results to third parties without consent. Fairlinked ...
Modern SQLite: Features You Didn't Know It Had
Modern SQLite (3.35+) ships JSON querying, FTS5 full-text search, window functions, strict typing, generated columns, and WAL mode, removing many external se...
Back to FreeBSD: Part 2 - Jails
FreeBSD jails provide OS-level isolation through one kernel primitive, unlike Linux containers combining namespaces, cgroups, and seccomp. Comparison with LXC.
Claude Code Unpacked
An interactive deep-dive into Claude Code's internals: the agent loop, 40+ tools, multi-agent orchestration, slash commands, mapped from the open source.
ngrok: AI & API Gateway - Secure Tunnels & Traffic
ngrok evolved from localhost tunneling into a cloud networking gateway for API routing, AI traffic, DDoS protection, and CEL-based Traffic Policy rules.
Ollama is now powered by MLX on Apple Silicon in preview
Ollama 0.18 now uses Apple MLX on Apple Silicon for faster local LLM inference, with NVFP4 quantization, better KV cache, and Qwen3.5-35B-A3B in preview.