LLM-Powered Bloomberg/ Visible Alpha / Factset Query Tool
Budget: $50 – $0 AUD
I need a proof-of-concept that lets me pose natural-language questions to Bloomberg and immediately get structured answers pulled from its presentations, filings, and transcripts—no manual document uploads, no screen scraping.
Phase 1 focuses on Bloomberg Visible Alpha and FactSet. Typical prompts include “What was company XYZ’s FY19 guidance?” or “Show the historical 1H vs 2H seasonality.” The system therefore has to understand and retrieve both financial-metric data and broader company-performance insights.
Please wire ChatGPT, Grok, or another comparable LLM into Bloomberg’s data layer, handle authentication, and design the retrieval pipeline so the model cites exactly where each figure or statement was sourced. A lightweight front end (CLI or simple web form) is fine for now; the priority is clean, accurate data extraction and clear answer formatting.
If someone has experience in doing it please let me know as only taking people who have done this before
Phase 1 focuses on Bloomberg Visible Alpha and FactSet. Typical prompts include “What was company XYZ’s FY19 guidance?” or “Show the historical 1H vs 2H seasonality.” The system therefore has to understand and retrieve both financial-metric data and broader company-performance insights.
Please wire ChatGPT, Grok, or another comparable LLM into Bloomberg’s data layer, handle authentication, and design the retrieval pipeline so the model cites exactly where each figure or statement was sourced. A lightweight front end (CLI or simple web form) is fine for now; the priority is clean, accurate data extraction and clear answer formatting.
If someone has experience in doing it please let me know as only taking people who have done this before