A personalized HN feed that learns from your favorites
Hi HN, I've been a daily user here for almost 15 years. Over that time, my interests have shifted. I find I'm now more interested in deep-dive technical posts and personal blogs than the big tech announcements that often dominate the front page. The "top" feed was starting to feel stale, and I was spending more time digging through "new". So, I decided to build what I wanted: a personalized "For You" feed. Link: https://hn.shaped.ai It's a simple concept: you log in with your normal HN credentials, and as you favorite stories, it learns what you're interested in and re-ranks the…
In plain words
A personalized Hacker News feed that learns from user favorites to re-rank stories. Users log in with their existing HN credentials, and the system tracks which stories they favorite to understand their interests, then surfaces more relevant content instead of relying on the default top or new rankings. Built by a long-time HN user frustrated with the static front page, it uses AI and modern web technologies to adapt the feed toward deep technical posts and personal blogs rather than major tech announcements.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi HN, I've been a daily user here for almost 15 years. Over that time, my interests have shifted. I find I'm now more interested in deep-dive technical posts and personal blogs than the big tech announcements that often dominate the front page. The "top" feed was starting to feel stale, and I was spending more time digging through "new". So, I decided to build what I wanted: a personalized "For You" feed. Link: https://hn.shaped.ai It's a simple concept: you log in with your normal HN credentials, and as you favorite stories, it learns what you're interested in and re-ranks the feed to show you more of that content. How it was built (a 2-day hackathon): The Client: I used an AI coding assistant (lovable.dev) to generate the initial React/Next.js client. It was surprisingly effective at getting a functional baseline up and running quickly. The Backend: Since HN's official API is read-only, I set up a lightweight Supabase backend. It uses edge functions to proxy login/voting requests to HN's unofficial API and a Postgres DB to cache posts and user events (favorites, etc.). The Personalization: The ranking is powered by my own company's platform, Shaped. It ingests the posts and your favorite events in real-time. The core of the ranking logic is a configurable formula. It's essentially the classic HN algorithm with a personalization term multiplied in: (item.score / score_penalty + content_similarity) / (time_decay) The content_similarity is calculated by comparing a post's text embedding to an embedding of your recent favorites. The best part is that you can actually play with the score_penalty in the UI to make the personalization stronger or weaker. This is very much a v1. I'd love to get your feedback. Does the personalization feel right? Any bugs? What's missing that would make you use it daily? Next on my list are things like collaborative filtering (once there's enough data!), semantic search, and a "similar stories" feature. Here's also a more detailed write-up about how it was built: https://www.shaped.ai/blog/building-a-hackernews-for-you-fee... Thanks for checking it out!
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, September 2025
the whole month →
- AS
Commerce · Sep 2025 · anycrap.shop
- TE
I made a built-from scratch Wayland Compositor to display any GUI app* in the terminal! I think there is a lot of unexplored potential in custom Wayland compositors, a lot of really cool things you can embed existing applications into! So, I started with embedding apps into the terminal because that is the easiest input/output (output is just utf-8 and I use the great `chafa` library for that, and I just read from stdin for the input). If you have any other ideas for cool Wayland compositors, let me know. I purposedly wrote 80% the app in Typescript to appeal to the most developers and…
Dev tools · Sep 2025 · github.com
- IR
Years ago I stumbled across a basic version of this concept and it stuck with me. I knew if I was ever going to take on such a project, it would need to be flawless, but without coding experience it was just another idea that would never happen. By the end of 2024, as AI coding tools exploded everywhere, I finally had a way to make it real. I started from zero knowledge and spent months collaborating with AI agents as a learning experience. Every pixel and every function went through me. The AI translated what I asked for into code, but every decision was human. I didn't use existing OS…
AI · Sep 2025 · mitchivin.com
