Bellman is a one-stop source for AI news. It functions as a news digest, pulling articles from trusted sources and curating them with AI as the central topic. It processes the article content, summarizes it, and provides key takeaways and detailed explanations through a chat interface.
+---------------------+
| Web Browser |
+---------------------+
|
| Streamlit App
|
+---------------------+
| Streamlit |
| Server |
+---------------------+
|
|
+---------------------+
| Snowflake Cortex |
| (Language Models) |
| |
| - Summarization |
| - Classification |
| - Chat |
+---------------------+
|
|
+------------------------+
| Snowflake Cloud |
| Database |
| |
| - Article Data |
| - User Preferences(TBD)|
+------------------------+
|
|
+---------------------+
| News API Sources |
+---------------------+
- Python
- Streamlit library (including experimental components like dialog boxes)
- Snowflake Arctic family models (4K context window model for summarization and chat, snowflake-arctic-embed-m embeddings model for classification)
- Snowflake Cortex
- Snowpark library
- Snowflake Cloud interface
-
Prepare virtual environment using tools like conda or venv.
-
Install dependencies
pip install -r requirements.txt
- Run the app
python -m streamlit run app.py
Used Streamlit for building the application interface Leveraged Snowflake Arctic family models for summarization, chat, and classification functionalities Interacted with Snowflake models through Snowflake Cortex and Snowpark library Implemented category filtering and card detail view with summary and insights
Pulled articles from a pool of trusted sources Processed article content using Snowflake models for summarization and classification
Streamlit community cloud was used for deployment
Keeping up with the influx of new terms, products, and companies in the rapidly evolving AI landscape can be overwhelming. Bellman aims to provide users with a comprehensive overview of the AI landscape, helping them stay updated with the latest trends with minimal stress.
Yumi Ko: Worked on project ideation, development plan, front-end development, database connection module, and app deployment.
Anthony Santana: Worked on app design, front-end development, API integration, and prompt engineering.