{"slug": "save-yourself-from-drowning", "title": "Save yourself from drowning", "summary": "A developer built and open-sourced an internal tool called Dev.to ASSISTANT that uses a Gemma 4 layer to summarize Dev.to posts by tag, extracting key details like project purpose, tools used, and engagement metrics. The tool aims to help builders quickly find high-value projects without scrolling through lengthy posts. The backend uses Python (FastAPI) and the frontend uses Next.js, with the entire project available on GitHub.", "body_md": "Dev.to is an absolute goldmine. On any given day, thousands of builders are sharing brilliant engineering lessons, launching side projects, breaking down new AI models, or submitting clever hackathon entries.\n\nThe problem is that this massive amount of information is hard to process. When a new tech challenge or hashtag trends, it is difficult to find the best work.\n\nWe want to support fellow builders, but nobody has the time to click through fifty long posts just to see what someone built and how they built it.\n\nI wanted a simple way to find high-value projects without spending my entire evening scrolling. So, I vibe-coded an internal tool called Dev.to ASSISTANT and decided to open-source it.\n\nInstead of explaining it, here is the tool turning the noise into a clean, structured summary in real-time:\n\nThe goal isn't to skip reading—it's to figure out exactly where to dive deep.\n\nWhen you pass the assistant a tag (like `#weekendchallenge`\n\nor `#ai`\n\n), it uses a **Gemma 4** layer to extract the meat of each post. You instantly get:\n\nThe exact \"why\" and \"how\" behind the project, stripped of fluff.\n\nWhat tools, frameworks, or models were used.\n\nLive reaction and comment counts so you can see what's sparking conversation.\n\nA link right back to the original Dev.to post so you can jump in and engage with the author.\n\nBecause Dev.to is an open ecosystem that generously provides an official API, I wanted to build this in that exact same spirit.\n\nThe backend is built with Python (FastAPI) and the frontend dashboard uses Next.js. I’ve made the entire project completely open-source so any builder can run it locally, tweak the filters, or add new curation features.\n\nIf you want to discover great engineering lessons faster, track challenge submissions, or just find amazing people to follow without getting lost in the feed, give it a spin.\n\n[Check out the GitHub repository here](https://github.com/siddqamar/dev.to-assistant) ⭐️\n\nIf this tool helps you find a cool repository or saves you some scroll time, I’d love to hear your thoughts in the comments below!", "url": "https://wpnews.pro/news/save-yourself-from-drowning", "canonical_source": "https://dev.to/qamar_dev_01/save-yourself-from-drowning-4e67", "published_at": "2026-07-09 08:12:16+00:00", "updated_at": "2026-07-09 08:41:16.276205+00:00", "lang": "en", "topics": ["developer-tools", "artificial-intelligence", "large-language-models"], "entities": ["Dev.to", "Gemma 4", "Python", "FastAPI", "Next.js", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/save-yourself-from-drowning", "markdown": "https://wpnews.pro/news/save-yourself-from-drowning.md", "text": "https://wpnews.pro/news/save-yourself-from-drowning.txt", "jsonld": "https://wpnews.pro/news/save-yourself-from-drowning.jsonld"}}