Affect.

• 

  News feed

• 

• 

Inspiration

Most financial news is written for Wall Street, not for the person paying rent in Mumbai or buying a car in New York. We noticed a massive "translation gap": a headline like "Central Bank raises repo rate by 25bps" sounds like noise to most people, but it actually means their monthly EMI just went up by ₹1,200. We built Affect to bridge that gap—turning cold global data into warm, personalized awareness.

What it does

Affect is a "Personal Economic Radar." It pulls live global business and economic news and immediately analyzes every story through the lens of your specific life stage. Using large language models, Affect calculates how a news story affects you based on your age, housing status (renter vs. owner), employment, and income. It then translates these stories into a "Cinematic Editorial" feed, ranking the news by the severity of its impact on your wallet.

How we built it

We built the frontend using Next.js 15 (App Router) to ensure a lightning-fast, SEO-optimized experience. For styling, we utilized Tailwind CSS v4 to create a cutting-edge design system built around a "dark luxury newspaper" aesthetic, complete with fluid Framer Motion animations.

The core intelligence is powered by Llama 3.3 70B (via the OpenRouter API), which handles complex economic reasoning securely via Next.js Server Actions. We pull real-time data using the NewsAPI, and designed a custom hook to create a seamless, responsive layout that shifts from mobile-app bottom navigation to a full desktop sidebar experience.

Challenges we ran into

The biggest challenge was balancing AI latency with user experience. Running intense LLM analysis on dozens of news articles simultaneously can be slow. We had to implement a strict server-side processing pipeline and intelligent caching to ensure the feed feels snappy and responsive without crashing the browser. Additionally, safely protecting API keys in a client-heavy web app required us to migrate our core AI logic into isolated, secure Server Actions while dealing with strict hydration boundaries.

Accomplishments that we're proud of

We are incredibly proud of the tone and identity of the app. Affect doesn't feel like a boring financial dashboard; it feels like reading a premium, beautiful editorial that happens to know you. We’re also proud of how meaningful the personalization is. Seeing the AI correctly tell a Renter that "House prices falling" is good news, while telling a Homeowner that it's cautionary, was a massive Eureka moment.

What we learned

Building for Hackanomics taught us the raw power of combining modern AI with dense macroeconomic data. We learned that the real challenge of "open finance" isn't just about accessing open datasets; it's about translation. We discovered how to successfully prompt large language models to act as personal actuaries rather than just summarizers. Navigating the complexity of transforming terrifying economic jargon into a simple, actionable "+$40/month" metric proved to be the ultimate Hackanomics lesson in bridging the gap between global economies and the individual wallet.

What's next for Affect

Our next big step is introducing "Pocket Event" push notifications, which will alert users only when a high-impact event (like a sudden local tax change) directly targets their specific demographic. We also want to integrate multiple hyper-local news sources and allow users to sync their actual monthly budgeting categories so the AI can provide "Cent-perfect" impact accuracy rather than estimates.

Built With

• api
• css
• llama
• newsapi
• next.js
• openrouter
• tailwind