Dottie Deeds
Applied AI for California Real Estate Document Automation
Dottie Deeds is an AI-powered legal workflow platform that helps attorneys and real estate professionals prepare California deed packages more efficiently.
The platform combines legal domain expertise, structured business logic, and large language models to automate repetitive document preparation while preserving professional review and judgment. Rather than replacing attorneys, Dottie is designed to eliminate administrative work so legal professionals can spend more time solving client problems.
Why I Built Dottie
After more than fifteen years practicing business and real estate law, I realized that preparing deeds and related transfer documents remained one of the most repetitive workflows in an otherwise sophisticated legal practice.
The legal analysis often required significant expertise.
The document preparation did not.
Modern AI created an opportunity to rethink that workflow from the ground up.
Dottie is the product I always wished existed while practicing law.
The Problem
Preparing California real estate transfer documents is still surprisingly manual.
Attorneys and paralegals routinely:
- Gather client and property information
- Determine the appropriate transfer mechanism
- Draft multiple recording instruments
- Verify statutory language
- Assemble county recording packages
- Produce Word and PDF deliverables
- Proofread formatting and execution requirements
While each transaction is unique, much of the document preparation follows repeatable patterns that can be significantly streamlined through software.
The Solution
Dottie combines structured legal workflows with AI-assisted document generation to reduce repetitive work while maintaining attorney oversight.
Current areas of focus include:
- Grant Deeds
- Quitclaim Deeds
- Interspousal Transfer Deeds
- Trust Transfers
- LLC and Entity Transfers
- Recording package preparation
- Intelligent document assembly
The objective is not to automate legal judgment.
The objective is to automate everything surrounding it.
Architecture
Frontend
- HTML
- JavaScript
- Responsive UI
Infrastructure
- Cloudflare Workers
- Supabase
- Vercel
- GitHub
AI
- Anthropic Claude API
- Prompt orchestration
- Structured document generation
Output
- Attorney-ready Microsoft Word documents
- Recording-ready PDF packages
- Standardized legal formatting
Product Principles
Every design decision is guided by five principles:
- AI should augment professionals, not replace them.
- Legal workflows require transparency and explainability.
- Security must be designed into the product from day one.
- Simplicity beats feature bloat.
- Real customer workflows should drive product decisions.
Current Status
🚧 Active Beta Development
Current capabilities include:
- Secure backend using Supabase
- User authentication
- Account management
- AI-powered document generation
- Cloudflare Worker API proxy
- Anthropic integration
- Usage monitoring and analytics
- Production deployment through Vercel
The current focus is onboarding beta users, measuring product usage, and iterating based on real-world feedback.
Roadmap
Near Term
- Expand supported California deed types
- Improve AI-assisted legal reasoning
- Streamline document assembly workflows
- Collect structured beta feedback
- Improve analytics and usage insights
Long Term
- Complete real estate transfer workflow automation
- Multi-document transaction support
- Firm administration tools
- Additional jurisdictions
- AI-powered transaction intelligence
Design Philosophy
I believe AI should augment professionals, not replace them.
The most valuable AI products are not those that generate the most text, but those that remove the most friction from real-world workflows.
Dottie Deeds reflects that philosophy.
Lessons Learned
Building Dottie reinforced that successful AI products require far more than connecting a language model to a user interface.
They require deep domain expertise, thoughtful workflow design, security, product judgment, and continuous iteration informed by customer feedback.
Those principles continue to shape every product I build.
About the Author
William B. Plevy is a California attorney, California real estate broker, and product builder focused on applying AI to complex professional workflows.
His work explores how legal expertise, product thinking, and modern AI can be combined to reduce operational friction in high-trust industries.
My Philosophy
I don't believe the future belongs to professionals who simply use AI.
I believe it belongs to professionals who understand a domain deeply enough to redesign it.
My work sits at the intersection of domain expertise, systems thinking, and applied AI.
I build software that removes friction from high-trust professional workflows, allowing experts to spend less time on repetitive process and more time on judgment, strategy, and human relationships.
I believe AI should augment professionals, not replace them.
The most valuable AI products are not those that generate the most text, but those that remove the most friction from real-world workflows.
Every product I build reflects that philosophy.
