Motivation and values
The next frontier in corporate document management is here. With the ever-increasing mass of documents in any organization, we aim to introduce an intelligent company knowledge assistant that seamlessly navigates the intricate relationships between diverse documents.

Questions like:
“Where can I find the right information for this task?”
“Where is the presentation I need located?”
“What was that detail I don’t remember about that contract?”
Will never take hours and hours to be answered.
A chatbot like this would save endless hours spent to document retrieval to any organization. This chatbot, powered by Retrieval Augmented Generation (RAG), will dive into SharePoint hosted documents, delivering answers in real-time and staying updated with immediate changes or additions to the documents.

Goals
Make this chatbot dream become true. Investigate the power of vector and graph database when it comes to augmenting the LLM. Deal with real time additions and deletion of the documents.

Methods
1. Vector and Graph Database Integration: Through advanced methodologies, SharePoint document data will be channeled into both a vector database and a graph database. This not only forms a rich, interconnected knowledge hub but also allows for ultra-fast information retrievals.
2 .Augmented LLM Deployment with RAG: By deploying state-of-the-art LLMs, which are further bolstered by the RAG technique, we'll ensure that these models can delve into the enriched datasets from the vector and graph databases, offering insightful, precise answers to queries.
3. Testing and Validation:
• LLM and Vectorization Techniques Comparison: Dive deep into multiple LLM variations and an array of vectorization techniques. This will ensure the selection of the most efficient and precise combination for the task at hand.
• Custom Query Set: A curated list of intricate questions targeting various document formats and content depths.
• KPI Metrics: Creation of KPIs focusing on response accuracy, speed, and contextual relevancy of the LLM answers.
• Extended Comparative Analysis: If time permits, delve into diverse data models for the graph database.
4. Cloud Deployment on AWS and Azure(if time permits it): Recognizing the importance of scalability and performance, our deployment will leverage the capabilities of AWS cloud
platforms. This cloud strategy will ensure data redundancy, high availability but also empower our model to provide real-time insights, adapting immediately to document alterations. If there is
time we will extend the project on Azure to explore multi-cloud features.

Requirements and opportunities
We don’t need any specific requirement from the students other than a great motivation for learning and proactivity in any of the project’s phases. The more you will invest on the project
the more you will get out of it. Engaging in this project offers students a rare glimpse into merging state-of-the-art technologies with practical enterprise challenges. It's an academic-corporate collaboration that stands to
redefine how organizations perceive and manage their knowledge repositories. We can’t wait to start!

Apply to this project here