AI copilot for engineering project tenders
NLP:
Objectives
Develop an AI copilot to help engineers submit succesful tenders based on previous projects.
Problem
TYPSA is a large engineering company whose engineers frequently need to produce documents to respond to tenders. These tender responses typically reference some of the many projects that TYPSA has done before. Retrieving relevant examples from the vast amounts of pre-existing documentation relating to transport, water, building, renewable energy, and rural development projects can be difficult and time consuming. The primary issues included:
- Inefficiency: Searching through a vast library of documents to find information relevant to new tenders was time consuming and inefficient.
- Inaccuracy: Manual searches could result in missed information or incorrect data interpretation.
- Fragmentation: Information was scattered across different formats and sources, making it difficult to access relevant data quickly.
Solution
To address these challenges, we implemented TYPSA - Tech Thinker, an advanced Artificial Intelligence (AI) driven solution designed to streamline information retrieval and enhance productivity. Key elements of the solution included:
- Custom GPT Model: We developed a tailored AI assistant that understands and processes TYPSA’s specific knowledge base, including PDFs, DOCs, DOCXs, and TXT files.
- Metadata Integration: We used metadata about the files to generate quick initial responses, ensuring accuracy and relevance.
- User-Friendly Interface: We created an intuitive interface using OpenAI’s custom GPT functionality for seamless interaction and efficient information retrieval.
- Automated Document Conversion: We implemented scripts to convert various document formats into TXT for uniform processing.
Impact
The implementation of TYPSA - Tech Thinker has led to significant positive outcomes:
- Enhanced Efficiency: Reduced the time spent on document searches, allowing employees to focus more on core tasks.
- Improved Accuracy: Provided precise answers based on a comprehensive understanding of TYPSA’s documentation, reducing errors and misinterpretations.
- Increased Productivity: Streamlined the information retrieval process, boosting overall productivity.
- Scalability: The solution can handle large volumes of data and complex queries, supporting TYPSA’s growing needs.
