Revax provides AI-powered research and drafting tools for professional advisors, leveraging large language models integrated with in-house knowledge. The platform generates client-ready deliverables, ensuring consistent technical positions and drafting styles across the firm. It facilitates seamless export of drafts, complete with diagrams, directly into firm-specific PowerPoint templates for efficient review.
Funding
Funding not disclosed
Founders
Product
Problem
Professional advisors face significant time and resource constraints in researching and drafting client-ready deliverables. The process of embedding and structuring in-house knowledge, retrieving relevant information from disparate sources, and generating initial drafts is often manual and inefficient.
Solution
Revax offers a platform that automates the research and drafting process for professional advisors by leveraging advanced large language models and a custom deep research agent. The system embeds and structures a firm's in-house knowledge, making it readily accessible. Its deep research agent then retrieves pertinent information from both internal repositories and external sources, including current legislation and web data. Frontier large language models generate initial drafts of client deliverables, which are then presented for human review. This streamlines the workflow, enhancing efficiency and reducing the time required to produce high-quality client documentation.
Target Audience
The primary target audience includes advisory firms and professional advisors who require efficient methods for researching and drafting client deliverables.
Features
- **Knowledge Embedding:** Connects to existing software suites (e.g., OneDrive, Dropbox, Google Drive, Notion, Jira) to embed and structure in-house knowledge.
- **Deep Research Agent:** Utilizes semantic search and metadata filtering to surface relevant precedent reports from embedded knowledge bases.
- **Reranking Layer:** Incorporates recency bias to prioritize the most current and authoritative information.
- **External Source Integration:** Extends search capabilities to current legislation and the web when internal knowledge is insufficient.
- **LLM-Powered Drafting:** Employs large language models from providers such as OpenAI, Anthropic, and Google to generate initial drafts.
- **Partner-Approved Wording Retrieval:** Focuses on finding and incorporating approved language and relevant legal frameworks.
- **Metadata Filtering:** Intelligently applies metadata to refine search results within internal knowledge bases.