DEV CommunityFriday · May 22, 2026FREE

That quote I half-remember at 2am? I can find it now.

ailocal-llmwhisperollamaproductivity

gemma-brief is a fully local AI pipeline built by a college student to solve the problem of keeping up with educational YouTube content. It monitors specified channels, downloads new videos overnight, transcribes them using OpenAI's Whisper, summarizes with Google's Gemma 4 E4B model running on Ollama, enriches summaries with Wikipedia context, and sends a formatted PDF to Telegram each morning. The system runs entirely on a base MacBook Air with no cloud dependencies or API costs. The creator highlights that the tool allows them to ask questions about content from weeks ago, effectively providing a searchable memory for video content. The project is open-source and aims to help others with limited time and budget stay informed.

// why it matters

Enables developers to stay current with video content without manual watching, using only local hardware.

Sources

Primary · DEV Community
▸ Read original at dev.to

Like this? Get the next digest.