This project is a real-time news sentiment platform built to turn fast-moving global news into structured sentiment data.

The idea behind it was simple: news moves quickly, but raw headlines are difficult to use inside dashboards, trading tools, AI workflows, or business intelligence systems without first being processed and organised. This platform takes news content and presents it through sentiment scores, topic-based datasets, historical views, alert concepts, and API access.

I built it as a product-style platform rather than just a landing page. It includes a real-time sentiment testing area, daily and weekly dataset previews, historical trend sections, API documentation, and developer-focused access through RapidAPI.

The project helped me think more deeply about building data products. It is not just about showing information on a page, but about turning live information into something structured, repeatable, and useful for other systems.

N/A

One of the main challenges was turning a broad AI idea into something clear and practical. Sentiment analysis can sound abstract, so the project needed to show what the data could actually be used for: testing headlines, viewing datasets, tracking topics, checking historical trends, and accessing structured results through an API.

Another challenge was making the platform feel like a real data product. The site needed to do more than explain the idea. It had to present live analysis, dataset previews, API access, and product use cases in a way that made sense to both technical and non-technical users.

The positioning was also important. I had to keep the language simple enough for business users, but still specific enough for developers, analysts, and people working with AI or data systems.

This project taught me that AI products need a clear structure around the output they create. The AI part is only useful when the result is understandable, consistent, and easy to use in a real workflow.

I also learned more about how to think in terms of datasets and APIs rather than just pages. A product like this needs repeatable data formats, clear access points, and a frontend that helps users understand what the system is doing.

It also pushed me to think more like a product builder. I had to consider the problem, the user, the data, the interface, and the way the platform would be positioned as a real tool rather than just a technical experiment.