I Vibe Coded an AI Anki Deck Generator From Scratch (and You Can Too) | Part 1

Ever wished you could generate a full Anki deck - with audio, images, and vocabulary - for any language topic in seconds? I built exactly that, live, with no pre-written code based on a spec file. This is a full vibe-coded session where I go from zero to a working Next.js app called Vagabond that uses AI to create immersive language flashcards. In this video you'll learn: - How Anki .apkg files actually work under the hood - How to generate vocabulary cards using an LLM (via OpenRouter) - How to add real audio with Google Text-to-Speech and contextual images with Gemini Imagen - How to vibe code a multi-step web app with Next.js and React - no boilerplate, just building Whether you're a language learner tired of making cards by hand, or a developer curious about AI-powered app building, this one's for you. By the end, you'll know how to build and customize your own Anki deck generator. GitHub repo: https://github.com/malkavian-librarian/vagabond Spec file: https://github.com/malkavian-librarian/vagabond/blob/master/specification.md #AnkiCards #VibeCoding #LanguageLearning #AIApps #NextJS #LearnToCode #FlashCards #SpacedRepetition #BuildWithAI #IndieHacker #LLM #GoogleGemini #OpenRouter #LanguageLearner #SideProject #antigravity