Technology
How it works, and what it does
System Architecture
High-level overview of Faxi's components and data flow
What the AI does
One faxed page in, the task carried out
Read, understand, act
Faxi reads the page, finds the marks made on it, works out what was asked for, and carries it out. One faxed page in, a finished task and a reply fax out.
Vision AI
Optical Character Recognition and Visual Analysis
Extracts text from fax images including both printed and handwritten content. Uses advanced computer vision to understand document structure, identify form fields, and recognize Japanese characters with high accuracy.
Annotation Detector
Visual Annotation Recognition
Identifies hand-drawn marks on faxes such as checkmarks, circles, arrows, and underlines. Associates annotations with nearby text to understand user intent.
Planner
Natural Language Understanding and Action Extraction
Analyzes extracted text and annotations to determine what action the user wants to perform. Classifies intents (email, shopping, appointment, etc.) and extracts relevant parameters with high confidence.
What happens to your fax
The fax arrives
We take the page exactly as it was sent. No special paper, no special machine.
We read it
Printed text and handwriting alike, including circles and checkmarks made on the page.
We work out the request
What was asked for, and what it needs. If anything is unclear, a fax goes back to confirm rather than a guess going forward.
We carry it out
Send the email, place the order, look something up, book the appointment.
The reply is faxed back
Results come back on paper, with a reference number so the next fax can pick up where this one left off.
Overall Performance
Technology Stack
Modern, scalable technologies powering the Faxi platform
Frontend
Astro
Fast static site generator for optimal performance
TypeScript
Type-safe JavaScript for robust code and better developer experience
Tailwind CSS
Utility-first CSS framework for rapid UI development
Backend
Express.js
Fast, minimalist web framework for Node.js
PostgreSQL
Robust relational database for storing users, jobs, and metrics
Redis
In-memory data store for job queues and caching
AI & Machine Learning
Multimodal AI
Reads the fax, understands the request, and plans the actions
Infrastructure
Cloud fax transport
Sends and receives faxes to Japanese and international numbers
Docker
Containerization for consistent development and production environments
MCP Integration
Model Context Protocol servers extend Faxi's capabilities
What is MCP?
Model Context Protocol (MCP) is an open standard that enables AI systems to securely connect with external data sources and tools. Faxi uses MCP servers to extend functionality beyond basic fax processing.
Email MCP Send and manage emails on behalf of users
Capabilities
- Send emails with attachments
- Reply to email threads
- Search and retrieve emails
- Manage email folders
Example Use Case
User faxes a handwritten letter → AI extracts content → Email sent to recipient
Shopping MCP Browse products and place orders
Capabilities
- Search product catalogs
- Add items to cart
- Process payments
- Track order status
Example Use Case
User circles products in catalog → AI detects selections → Order placed automatically
AI Chat MCP Enable conversational AI via fax
Capabilities
- Answer questions
- Provide recommendations
- Explain complex topics
- Multi-turn conversations
Example Use Case
User writes question on fax → AI generates response → Answer faxed back
Payment MCP Process payments and manage transactions
Capabilities
- Register payment methods
- Process bill payments
- Issue payment codes
- Transaction history
Example Use Case
User faxes utility bill → AI extracts amount → Payment processed
Appointment MCP Schedule and manage appointments
Capabilities
- Check availability
- Book appointments
- Send reminders
- Reschedule or cancel
Example Use Case
User requests doctor appointment → AI finds slots → Appointment confirmed