Investor Relations AI Assistant — Arabian GAS
Client
Arabian GAS
Industry
Investor Relations
Year
2025
The Challenge
Arabian GAS needed to give investors fast, accurate answers drawn from a large and constantly growing body of corporate knowledge — annual reports, official disclosures, PDFs, and website content — without overloading the investor relations team or retraining a model every time information changed.
Our Solution
We built a specialized investor relations assistant powered by Retrieval Augmented Generation (RAG), allowing the system to reach institutional knowledge across multiple sources and deliver precise, up-to-date answers without model retraining. The assistant understands investor questions in Arabic or English and searches across annual reports, official disclosures, PDF files, and published website content.
- Bilingual support (Arabic and English)
- Search across thousands of pages and documents
- Accurate, trustworthy answers
- Easy updating of knowledge sources
- A natural, fast, and interactive experience
Results
Reduced workload on the investor relations team
Faster access to accurate information
Consistent, source-grounded responses for investors
- Reduced workload on the investor relations team
- Faster access to accurate information
- Consistent, source-grounded responses for investors
Technologies Used
- Retrieval Augmented Generation (RAG)
- Vector search and semantic retrieval
- Large language models with bilingual NLP
- Document ingestion pipeline (reports, disclosures, PDFs, web)