Better Stack - Videos
Back to ChannelSpec Kit: Github's NEW tool That FINALLY Fixes AI Coding
Spec-driven development is here to change the way we code. In this video, we break down GitHub’s new open-source toolkit, Spec Kit, and show how it transforms messy vibe coding into structured, rel...
Poml: Microsoft's Secret Tool for PERFECT AI Prompts
Poml is Microsoft's markup language for crafting better AI prompts. In this video w go through how Poml transforms HTML-like code into powerful, structured prompts for any LLM, complete with data i...
This Voice AI Creates Full Podcasts in Minutes! (Microsoft's VibeVoice)
Microsoft just released VibeVoice, a powerful new open-source AI that can generate entire podcasts and audiobooks with up to four realistic voices. In this video, We’ll walk you through how VibeVoi...
Codex CLI Has Just Got WAY Better!
Codex CLI from ChatGPT creators OpenAI, has just levelled up. In this video, I show why the Rust rewrite makes it better, how the minimal UI keeps you focused, and what GPT‑5 High unlocks for codin...
The BEST Way to Manage Versions of Node, Python, Go (and Much More...)
Mise-en-place or Mise a polygot package manager, streamlines developer onboarding by installing and managing packages like Node, Python, Go, and even NPM packages from a single TOML file. In this v...
SDKs, Stealth Models, and LLM Myths | Better Stack Podcast Ep. 2
In this episode of the Better Stack Podcast, the team dives into the latest developments in AI tooling and models. They explore the new AI SDK and AI Elements library, discuss the stealth release o...
Bun.js Just Got WAY Faster (1.2.21 Update)
Bun.js just released versions 1.2.20 and 1.2.21 with big improvements. In this video, we’ll break down the new bun.lock file, faster AbortSignal timeout, updates to bun:test, and the all-new unifie...
INK: The React Library POWERING AI Terminal Tools (For Now...)
Claude Code, Gemini CLI and Qwen Code have one thing in common, they are all built in React + TypeScript using Ink, but it's not perfect.. In this video we go through why so many AI command-line to...
How to run any Hugging Face model on your own server? #huggingface #ai
NVIDIA Quietly Dropped This POWERFUL AI Model (Nemotron v2)
NVIDIA quietly dropped Nemotron v2, a 9B parameter model that blends transformers with Mamba-2 for blazing fast reasoning. In this video I’ll break down how it works, why it’s different, and run re...
The Cursor CLI is Actually Good!?
I was skeptical about Cursor CLI at first, but after actually using it, I had to completely redo this video. The Cursor Agent's capabilities genuinely surprised me, even when using the same model a...
How does an LLM ACTUALLY Work? (Visual Breakdown)
Large language models like GPT-4, Claude, and DeepSeek might feel magical, but at their core they’re just predicting the next token in a sequence. In this video, we break down how transformers real...
The EASIEST way to build AI apps – AI SDK Overview
Build a complete AI-powered app in TypeScript in just 10 minutes using the AI SDK! This comprehensive tutorial covers everything from basic text generation to advanced features like streaming, stru...
The BEST Way To Use Claude Code Anywhere (Thanks To Levelsio)
Turn your mobile phone into a vibe coding machine. In this guide, I show how to install Claude Code on an always-on VPS, then connect from your phone using Termux or Termius—so you can code anywher...
We Found the BEST Vector Database! (Testing Head-to-Head Benchmarks)
In this video, we build a semantic music search engine and benchmark five vector databases in a real app. We compare Qdrant, Milvus, Weaviate, Pinecone, and TopK on speed and accuracy with clear, s...
Is Stealth Grok 4 Code Model (Sonic) Actually Good?
Sonic AI or Grok 4’s coding model—just dropped, promising fast reasoning, a 262k context window, and real dev chops. In this hands-on test, we put Sonic to the test building: a React + Three.js 3D ...
Better Stack Podcast Ep 1: The GPT-5 Backlash, AI CLI Tools & AI Agent Protocols
Join James, Richard, and Andris as they dive deep into the explosive ChatGPT-5 release that left developers fuming! Discover why the "PhD-level coding AI" fell flat in real-world tests, the controv...
Claude Code Just Got Way Better (Here’s How)
Claude Code has been updating so frequently that there are some things you might have missed. This quick walkthrough highlights the most useful recent upgrades: custom status lines for instant cont...
AI Agent Protocols Explained: MCP, A2A, ACP and More
AI agents are opening new possibilities for automation, but the protocols behind them can be tricky to navigate. In this video, we will walk you through MCP, A2A, ACP, and ANP with practical demos ...
The HIDDEN Reason GPT‑5 Feels Underwhelming
Why does GPT‑5 sometimes feel underwhelming even though it tops benchmarks? In this video I explain the hidden reason: OpenAI’s real‑time routing system that dynamically picks smaller or larger sub...
Add Social Logins to Next.js in 5 Minutes (Supabase UI)
Adding social logins to a Next.js app has never been this easy. In this video, I’ll show you how to add social logins/authentication using Supabase UI — with just ONE command 🔗 Relevant Links http...
Why Crush Is the Best‑Looking AI Coding CLI (RIP Claude Code?)
Charm's Crush proves that terminal tools can be beautiful. In this video, I dive into Charm’s design language and show why Crush feels premium: from its clean headers and logo, to smooth loading an...
Shadcn's AI Elements are AWESOME (easiest way to build a chat app)
Build beautiful AI chat apps in minutes with shadcn's new AI Elements! This component "library" from Vercel makes creating ChatGPT-style interfaces incredibly simple - complete with markdown render...
Your Own Mini AWS in Under 30 Minutes (Incus Tutorial)
Incus is a powerful open-source tool for running containers and virtual machines, and in this video we’ll set it up from scratch. You’ll learn how to install Incus, configure networking, run Ollama...
Warp + Neovim + Superfile: This Terminal Setup CHANGED How I Code
Ditch the mouse and supercharge your workflow with my full terminal-first setup: Warp Terminal + Fish shell + Neovim and more. In this video, I share how I went from VS Code to living in the termin...
Build an AI Video Generator Like Sora (with ComfyUI)
In this video, we walk through building our own AI text-to-video generator using ComfyUI, from setting up a GPU server to integrating it into a web interface. We explore how to combine text-to-imag...
GPT-5 Just Killed Claude – Smarter & Cheaper
🚀 OpenAI just dropped GPT-5 and it's a GAME CHANGER! With 5 new models, 8x cheaper pricing than Claude, and near-perfect hallucination fixes, this could finally pull developers away from Anthropic....
TypeScript 5.9: Lazy Imports Are Finally Here (and more)
🚀 TypeScript 5.9 is here with some seriously helpful improvements — from lazy module loading with import defer, to a much cleaner tsconfig setup, to a new --module node20 mode for more predictable ...
OpenAI FINALLY Goes Open... 2 NEW models (they're good?)
OpenAI just dropped their first open-weight models since GPT-2. GPT-OSS-120B and GPT-OSS-20B, both under the Apache 2 license. In this video, I run the 20B model locally, put both through real codi...
Everything New in Nuxt 4 — Should You Switch?
Nuxt 4 is finally here! 🎉 In this video, we break down everything that’s new, what’s changed, and whether it’s worth upgrading your existing projects. 🔗 Relevant Links Nuxt 4.0: https://nuxt.com/b...
TanStack Router is AWESOME #reactjs #javascript
Actually Useful MCP Servers #claudecode #ai
Why I Love TanStack Router
TanStack Router is the ultimate type-safe routing solution that's taking the TypeScript ecosystem by storm. From full type safety and prefetching to seamless data loading integration with TanStack ...
I Tested Google's Opal So You Don’t Have To #ai #vibecoding #nocode
Stack Overflow for your LLMs (and you!)
I found the PERFECT replacement for Stack Overflow and it's completely changed how I code. grep.app lets you search millions of GitHub repositories in milliseconds, giving you real code examples in...
Goose + Qwen3 Is INSANE — Best AI Tool I’ve Tried So Far
In this video, we test Goose, an open source local AI coding agent, paired with the powerful Qwen3-coder model. We walk through building real projects using features like hand gesture detection and...
Hurl: The Small Postman Killer You've NEVER Heard Of
Hurl is a lightweight, Rust-based HTTP runner that's quietly replacing Postman for developers who prefer working in the terminal. In this video, I'll show you how this simple CLI tool can handle AP...
The BEST Way To Start a Project (Better-T-Stack)
Stop wasting hours setting up your next project! I discovered Better T Stack - the ultimate tool that scaffolds modern TypeScript projects with ANY framework combination in just 30 seconds. It hand...
We Tested Google’s New AI Coding Tool Opal (Honest Review)
In this video, we take a deep dive into Google Opal, the new experimental no-code tool that lets you build AI-powered mini-apps using natural language and visual workflows. We put it to the test to...
Build Apps Using ONLY your Voice (WARP 2.0 is Insane)
🚀 Warp 2.0 is HERE! and it’s not just a terminal anymore. In this video, I’ll show you how Warp 2.0 can code full-stack apps, manage multiple AI agents, deploy to apps, and even take voice commands...
Qwen3 Coder vs Kimi K2 vs Claude Sonnet 4
Qwen3 Coder is out and boasting some impressive benchmarks. So in this video I put three top AI coding models to the ultimate test: building a real Next.js app that converts YouTube videos into blo...
DNS Malware Explained: How Hackers Hide Code in Plain Sight
Hackers are turning DNS into a secret malware storage system. In this video we explain how malware can be hidden inside DNS TXT records, why it is so hard to detect, and what you can do to protect ...
This One Claude Code Feature Replaced My ENTIRE Team
Anthropic just dropped a MASSIVE update to Claude Code. Custom Sub agents that will change how you code forever! In this video, I'll show you exactly how to create specialized AI agents for specifi...
Lynx: TikTok’s Blazing Fast Framework – The Next React Native Killer?
Lynx is ByteDance’s new open-source cross-platform framework, and many are calling it a potential React Native killer. In this video, we test Lynx, explore its dual-threaded architecture, and see h...
10 Chrome DevTools Tricks You Didn't Know! (3rd is Crazy)
In this video, I’ll show you 10 hidden DevTools tricks that I'm sure you didn't know. From enabling design mode to faking API responses with Local Overrides, simulating slow internet, debugging ani...
Task Master AI Has Changed How I Code FOREVER (w/ Claude Code)
Task master AI has completely changed my approach to AI coding with Claude Code. After you see what it built in ONE SHOT, you'll understand why this changes everything. In this video, I show you ex...
GitHub Spark Is INSANE – A Full App in Just 20 Minutes!
GitHub just released Spark, an all-in-one web-based platform for building apps with natural language prompts. Unlike traditional AI code assistants, Spark handles everything – from generating React...
Vue 3.6 ALPHA is out! and it's FAST! — Vapor
Vue 3.6 alpha is finally out! And it comes with a new feature: Vapor mode. It is a new compilation mode for Vue Single-File Components (SFC) with the goal of reducing baseline bundle size and impro...
Can AI Agents Finally Automate the Web? (BrowserOS Review)
BrowserOS is an open-source agentic browser that promises AI-powered web automation with full privacy control. In this video, we take a closer look at BrowserOS, test its key features, and show you...
The ULTIMATE Arch Linux + Hyprland Dev Setup – Omarchy
DHH's Omarchy just made Arch Linux + Hyperland setup ridiculously easy with a single script install. I tested it on an old bricked laptop and discovered 5 game-changing features that make this the ...