Skip to main content

GoodNews – A Brighter Way to Stay Informed

A refreshing take on the daily news cycle

GoodNews – A Brighter Way to Stay Informed
GoodNews Interface Example
GoodNews Interface Example

The Problem: News Doesn't Have to Be Exhausting

Most news platforms focus on conflict, disasters, and negativity. For many people, this creates stress and makes them avoid staying informed altogether. My goal with GoodNews was simple: what if staying updated could feel hopeful and bring actual joy instead of bringing out more anxiety? It is not a replacement to current news sources just an aggregator of positive news to highlight things that are overshadowed by negative news.

Contrast between traditional news and GoodNews
Contrast between traditional news and GoodNews

What I Built

GoodNews is a web app that uses AI to find and share uplifting stories from trusted sources. It focuses on progress, solutions, and underreported positive events—like breakthroughs in science or community-driven projects.

Key features:

  • Personalized feed: Pick topics you care about (e.g., climate wins, education).
  • Quick summaries: Get the key points in 4 sentences.
  • Mood settings: Switch between "Relaxed," "Inspired," or "Balanced" modes.
  • Transparency dashboard: See where stories come from and how they're verified.
Users adjust their feed based on how they want to feel
Users adjust their feed based on how they want to feel

How It Works

Personalized:

  1. AI filters stories: The app scans 100+ news sources and filters negative or sensational content. Every minute.
  2. Stories are embedded: A story summary gets generated and embedded, so that users can preview and can find relevant stories for them.
  3. You choose the topics: Write what you want to know about in natural language, up to 5 topics.
  4. Enjoy Good News: Once you've chosen the topics, those are the news you see, and you can change them whenever.
Topic selection interface
Topic selection interface

Behind the Scenes

Building GoodNews required:

  • Sentiment analysis tools to detect uplifting language.
  • Choose your topics so that everyone gets exactly what they want.
  • Rag based matching to topics using query expansion to have the most semantically correct matches.
Query expansion for semantic matching
Query expansion for semantic matching
  • Metrics of topics of interest and positivity to see where you got joy from and figure out new topics that may interest you.
  • No data collected, ever to signup you just get a uuid that you need to remember so that your topics stay the same, no email, no phone, just good news.
No tracking login screen
No tracking login screen

What I Learned

Not everyone wants positivity: A lot of people who tried it went back to their original news sources, not because they didn't like the content they were seeing, but because the anxiety that is commonly introduced in the normal news media is the thing that makes it so addicting.

User engagement metrics
User engagement metrics

The streamlined navigation keeps the focus on content that matters.

Navigation bar of GoodNews
Navigation bar of GoodNews

Content Categories

Users can browse specific topics like sports to see only positive stories in their areas of interest.

Sports category view
Sports category view

Tools Used

Figma, React, Python (for sentiment analysis prototypes), OpenAI

This project reminded me that technology can shape not just what we know, but how we feel. Designing for positivity isn't about ignoring problems—it's about highlighting the ways people are solving them.