Alexandria

Alexandria turns any URL into an interactive, voice-based lesson within seconds. From blog posts and YouTube videos to scientific papers and PDFs, the web is full of valuable content — but it's rarely optimized for learning.

With Alexandria, users can transform any link into a personalized educational experience, tailored to their language and knowledge level (from a 5 year old to a PHD graduate). The system lets you engage with the content through natural conversation, ask questions, and reinforce understanding through dialogue. To make learning even more impactful, Alexandria also supports token-based incentives — rewarding users for mastering the material and encouraging consistent progress.

We integrate with ElevenLabs to power the voice-based agent, feeding it the extracted context so it can respond naturally and guide the learner through the material. In parallel, we generate a dynamic set of flashcards based on the source content. These flashcards are surfaced client-side where we use ElevenLabs' tool-calling capabilities to let the voice agent control frontend interactions — highlighting flashcards, triggering animations, or emphasizing key concepts as it speaks.

This setup makes the agent feel more alive, more helpful — and ultimately creates a learning experience that’s both intelligent and intuitive.