Bisv Hacks

Published: at 05:59 AM

Bisv Hacks 2025

The Challenge

Using all modern technological tools and resources available to you, create innovative solutions that address issues of social justice and promote social good. Your challenge is to tackle real-world problems faced by marginalized groups and disadvantaged communities. You can code software, or build physical projects.

This open-ended challenge invites you to turn your ideas into impactful solutions, with countless opportunities to empower, support, and uplift those in need.

The Beginning

Bisv Hacks 2025 was an incredible experience, bringing together some of the brightest young minds in tech. The theme for the hackathon was “Innovating for a Better Future,” and we knew we had to come up with something unique and impactful to stand out. Our team, consisting of Me, Alex Liao, Sanjana Gowda, Shely Jain, and Vaidehi Akbari, was determined to create a solution that not only showcased our technical skills but also addressed a real-world problem.

Brainstorming the Idea

We were a little late with getting started on the project so we only had 12 hours.

With only 12 hours to brainstorm, design, and develop our project, we dove straight into idea generation. After several discussions, we decided to build a browser extension that detects political bias in news articles—something we realized could be a game-changer in the current media landscape.

The idea stemmed from the growing concerns about misinformation, biased reporting, and echo chambers in digital news consumption. Our goal was to provide users with an unbiased assessment of the content they read, helping them develop a more balanced perspective.

Development Process

Tech Stack & Challenges

We used:

One of the biggest challenges we faced was extracting meaningful text from various websites while avoiding unnecessary HTML elements like ads and sidebars. Using BeautifulSoup, we efficiently scraped and cleaned the text before sending it to the AI for analysis.

Another hurdle was integrating OpenRouter AI to analyze news articles accurately. Initially, we ran into API errors and inconsistent responses, but after refining our request format and improving error handling, we were able to get reliable results.

The Final Product

After hours of debugging and optimizing, we finally had a functional prototype! Our extension would:

  1. Automatically analyze the news article the user was reading.
  2. Provide a bias score and explanation, highlighting specific phrases that contributed to the bias.

Takeaways & Next Steps

Winning second place was a huge achievement, but more importantly, the experience taught us valuable lessons: