AI Financial Feeds: Real-time sentiment analysis for financial news

CLIENT
Internal project
WEBSITE
pragmaticcoders.github.io/ai-rss-feed/
Tools used
Make, Airtable, ChatGPT, JavaScript, HTML, Bulma CSS
MAIN SERVICE
AI workflow
Scroll to see more
AI Agent Financial Feeds: Real-Time Sentiment Analysis for Financial News

What did we create?

AI Financial Feeds by Pragmatic Coders acts as a smart, sentiment-aware news dashboard for financial markets, automatically categorizing and summarizing market-related news in a user-friendly way.

Key Features

  • News Aggregation: The tool pulls in stories (e.g., about company stocks, currency pairs, and broader financial events).
  • AI Sentiment Indicators: Each story is automatically labeled with ā€œBullish,ā€ ā€œNeutral,ā€ or ā€œBearishā€ to indicate how the news might affect market sentiment.
  • Impact Level: Stories are marked with a ā€œLow,ā€ ā€œMedium,ā€ or ā€œHighā€ impact label, showing how significant the news could be for markets.
  • Assets Overview: The tool summarizes overall sentiment for different asset typesā€”like stocks, USD/MXN, and USD/CADā€”based on the aggregated news.
  • Chat with AI Assistant: There is a built-in assistant feature that can respond to queries or provide more context about the news items.

Process

AI Financial Feeds make process

  1. Fetch News Items: The system automatically retrieves new articles or posts from an RSS feed whenever they appear.
  2. Clean and Format: The raw HTML is scrubbed of unnecessary tags (like scripts or ads), and the remaining content is converted to Markdown for easier processing.
  3. Prepare Data: Key details (title, content, source) are extracted and saved to variables so that each article is ready for analysis.
  4. AI Analysis: An AI model (e.g., ChatGPT – we chose GPT-4o due to its easy integration and fast response time ) classifies the article, summarizes it, determines sentiment (ā€œBullish,ā€ ā€œNeutral,ā€ ā€œBearishā€), and assigns an impact level (ā€œLow,ā€ ā€œMedium,ā€ ā€œHighā€).
  5. Categorize by Asset Type: Based on the AIā€™s analysis, articles are routed into different categoriesā€”such as stocks, forex, cryptoā€”to reflect which market sector each article impacts.
  6. Store Results: For each category, final data (including sentiment and impact scores) is aggregated and saved into a database (e.g., Airtable) for easy reference and further analysis.

Challenges

As always with AI solutions, the main challenge proved to be correctly gathering and formatting the data so that it could be processed by AI algorithms. A particularly critical aspect was the scraping and conversion of HTML content to Markdown, which required precisely defining the necessary data elements and extracting them.

Benefits

With AI Financial Feeds, users can quickly scan the site to see which assets or market sectors are seeing the most positive or negative press.

It helps traders, analysts, or anyone interested in finance keep a pulse on the latest market-moving headlines and gauge sentiment shifts as they happen.

AI Agent Financial Feeds: Real-Time Sentiment Analysis for Financial News

Contents

Let's talk

We’ve got answers on anything connected with software development.

Ask a question

You can ask us a question using the form below. We will respond as soon as possible.

Schedule a meeting

You can also schedule an online meeting with Wojciech, our Senior Business Consultant.

wojciech knizewski portrait
Wojciech Kniżewski

Senior Business Consultant

8 out of 10

founders who contacted us wanted
to work with our team.

Trusted partner

Newsletter

You are just one click away from receiving our 1-min business newsletter. Get insights on product management, product design, Agile, fintech, digital health, and AI.

LOOK INSIDE

Pragmatic times Newsletter