We are looking for a skilled professional to develop a system for training ChatGPT or Google BARD on company-specific documents to efficiently respond to questions. The primary goal is to design and implement a solution that can handle lengthy documents and accurately generate relevant responses. The project involves the following tasks:
- Develop a training pipeline to preprocess and extract relevant information from company-specific documents that are in the form of PDFs or Word files.
- Implement an interface to collect and label question-response pairs
- Train ChatGPT or Google BARD using the collected data
- Test and evaluate the trained model for accuracy and performance
Relevant skills:
- Natural Language Processing (NLP)
- Machine Learning
- Data preprocessing
- Text classification
This job is already closed and no longer accepting applicants, sorry.
