Stop Searching.
Start Knowing.
Your HR policies, technical SOPs, and sales proposals are scattered across Drive and SharePoint. Novus brings them into one secure chat interface with exact citations.
The "Secret Sauce"
We don't just dump text into the AI. We implemented advanced Parent-Child Chunking and metadata filtering. This allows the model to understand the full context of a 50-page complex contract, reducing hallucinations by over 60%.
Instant Answer & Citation
Don't search for "Termination Clause". Just ask.
Data Privacy Silos
SOPs and Contracts stay private. No training.
Upload Documentation
PDF or Markdown
Under the Hood
Parent-Child Chunking
We preserve context by indexing parent documents alongside chunks, ensuring LLMs have the full picture. Reduces hallucinations by 60%.
Secure Data Silos
Your vectors live in Pinecone/ChromaDB with strict metadata filtering. Essential for privacy-focused enterprise deployment.
Advanced Orchestration
Built using LangChain for sophisticated query routing and retrieval. Using GPT-4o or Claude 3.5 Sonnet for reasoning.
Frequently Asked Questions
Everything you need to know about the product and billing.
What is Novus?
Novus is an enterprise AI knowledge base that allows employees to securely chat with internal documents like contracts, SOPs, and HR policies using advanced RAG (Retrieval-Augmented Generation) technology.
Is my data secure with Novus?
Yes. Novus uses a privacy-first architecture. Your documents are indexed into a secure vector store (Pinecone) with strict session-based access controls. Your data is never used to train public AI models.
How to chat with company documents using AI?
Simply upload your PDF or Markdown documents to Novus. Our AI instantly indexes the content using vector embeddings. You can then ask questions in natural language and get accurate answers with exact citations from your documents.
What is RAG technology?
RAG (Retrieval-Augmented Generation) is an AI technique that combines large language models with a retrieval system. Instead of relying on the AI's training data, RAG retrieves relevant context from your documents before generating answers, resulting in more accurate and hallucination-free responses.
What is Parent-Child Chunking?
Parent-Child Chunking is an advanced document processing technique. We index smaller child chunks for precise retrieval but feed the larger parent document context to the AI. This significantly reduces hallucinations by ensuring the model understands full context, not just fragments.
Can I deploy Novus on-premise?
Yes. Novus is designed for hybrid deployment. While we use cloud-based LLMs (GPT-4o, Claude 3.5), the vector store and orchestration layer can be deployed within your private VPC for maximum security.
What's the best AI tool for contract review?
Novus is specifically designed for enterprise contract review. Our Parent-Child Chunking reduces hallucinations by 60%, and our citation engine provides exact source references so legal teams can verify every answer instantly.
Is Novus a private ChatGPT alternative for business?
Yes. Unlike public ChatGPT, Novus creates isolated sessions for each user. Your sensitive documents never leave your secure environment and are never used to train public models. It's the privacy-first alternative enterprises need.
Ready to Ship?
Join thousands of developers building the next generation of AI-powered internal tools.