Frequently Asked Questions
Find answers to common questions about Archevi. Can't find what you're looking for? Contact us.
AI Chat & RAG Queries
How do I ask questions about my documents?
Simply type your question in natural language in the chat interface. Archevi will:
- Search your documents for relevant information
- Synthesize an answer from multiple sources
- Cite which documents the answer came from
Example questions:
- "When does my car insurance expire?"
- "What's Grandma's apple pie recipe?"
- "Show me all medical records for 2024"
How does the AI agent work?
The RAG (Retrieval-Augmented Generation) agent:
- Analyzes your question to determine search strategy
- Searches documents using semantic similarity
- Can also search PDF pages for visual content (charts, signatures)
- Generates a response grounded in your actual documents
- Never makes up information not in your documents
Why does the AI sometimes not find information I know exists?
Several reasons this might happen:
- The document hasn't been embedded yet (check processing status)
- The question uses different terminology than the document
- The document is an image without OCR text extraction
- Try rephrasing your question or being more specific
How do I retrieve past conversations?
Click the chat history icon to see previous sessions. Each session includes:
- All messages exchanged
- Source documents cited
- Timestamps
Can I use voice to ask questions?
Yes! Archevi supports voice input:
- Click the microphone icon to record
- Audio is transcribed using Groq Whisper
- Responses can be read aloud using ElevenLabs TTS
Administration
How do I view admin activity logs?
Security audit logs track:
- Tenant configuration changes
- User management actions
- Backup operations
- API key usage
How do I manage database backups?
Backups are automatic, but you can:
- View available backups
- Trigger manual backup
- Restore from backup (with confirmation)
How do I list all families/tenants?
Admin endpoint for multi-tenant management:
Returns all families with member counts, document counts, and plan info.
Analytics & Usage Tracking
How do I monitor usage and quotas?
The Usage dashboard shows:
- Storage used vs limit
- AI tokens consumed this month
- Queries made vs limit
- Per-member breakdown
How do I analyze query patterns?
Query analytics show:
- Most common questions asked
- Average response times
- Error rates by type
- Model usage breakdown (Groq vs Cohere)
How do I track API costs?
Cost tracking by provider:
- Cohere (embeddings + reranking)
- Groq (chat completions)
- ElevenLabs (text-to-speech)
Breakdown by day, operation type, and tenant.
How do I forecast future costs?
Cost projections based on:
- Historical usage trends
- Growth rate analysis
- Budget alerts when approaching limits
Authentication & Security
How do I log in?
POST to the login endpoint with email and password:
Returns JWT token valid for 24 hours.
How do users reset their passwords?
- Click "Forgot Password" on login
- Enter email address
- Receive reset link (valid 1 hour)
- Set new password
How do I enable two-factor authentication?
- Go to Settings > Security
- Click "Enable 2FA"
- Scan QR code with authenticator app
- Enter verification code
- Save backup codes securely
What happens if I lose my 2FA device?
Use one of your backup codes to log in, then:
- Disable 2FA
- Re-enable with new device
Contact support if you've lost backup codes.
Billing & Pricing
What plans are available?
| Plan | Price | Storage | AI Queries | Members |
|---|---|---|---|---|
| Free | $0/mo | 1 GB | 50/month | 2 |
| Family | $9/mo | 10 GB | 500/month | 5 |
| Family+ | $19/mo | 50 GB | Unlimited | 10 |
| Family Office | $49/mo | Unlimited | Unlimited | 25 |
How do I upgrade my plan?
- Go to Settings > Billing
- Select new plan
- Enter payment details
- Upgrade is instant
What happens if I exceed my limits?
- Storage: Can't upload new documents
- AI queries: Queries pause until next month
- Upgrade anytime to resume immediately
Document Management
How do I search documents semantically?
Use the search bar or chat to search by meaning, not just keywords. Archevi uses AI embeddings to find documents that are conceptually related to your query.
Examples:
- "life insurance" will find "term policy" and "coverage documents"
- "tax documents" finds "W-2", "1099", and "tax return"
How do I filter searches by category or date?
Use advanced search to combine:
- Categories: Insurance, Medical, Financial, Legal, etc.
- Date range: Created/expiry dates
- Tags: Custom labels you've applied
- Family member: Documents assigned to specific people
How do I find documents expiring soon?
The Expiring Documents view shows:
- Documents expiring in the next 7/30/90 days
- Grouped by urgency (critical, soon, upcoming)
- Automatic notifications via email
How does document versioning work?
Archevi maintains version history for documents:
- Every edit creates a new version
- View previous versions anytime
- Rollback to any previous version
- See who made changes and when
How do I bulk upload documents?
Several options:
- ZIP upload: Upload a ZIP file with multiple documents
- Email forwarding: Forward emails with attachments to your family inbox
- API: Use the bulk upload endpoint for programmatic access
How does autocomplete work?
As you type in search, Archevi suggests:
- Document titles matching your query
- Tags you've used
- Family member names
- Recent successful searches
Family & Tenant Management
How do I invite family members?
- Go to Settings > Family Members
- Click "Invite Member"
- Enter their email address
- Select their role (Admin, Member, or Viewer)
- They'll receive an email invitation
What are the different roles?
| Role | Permissions |
|---|---|
| Owner | Full access, billing, can delete family |
| Admin | Manage members, all documents, settings |
| Member | Upload/edit own documents, view shared |
| Viewer | View-only access to shared documents |
How do I customize the look and feel?
Each family can customize:
- Logo and favicon
- Primary and accent colors
- Font family
- Theme (light/dark/system)
How do I switch between families?
If you belong to multiple families:
- Click your profile icon
- Select "Switch Family"
- Choose from your available families
Getting Started
What is Archevi?
Archevi is an AI-powered family knowledge base that helps families organize, search, and understand their important documents. It uses advanced AI to extract information, answer questions, and even generate biographies from your stored documents.
Key features:
- Semantic document search using AI embeddings
- Natural language Q&A about your documents
- Automatic expiry tracking for insurance, licenses, etc.
- Family timeline generation
- Secure document sharing
- Knowledge refinement through user confirmation
How do I get started with Archevi?
- Sign up for an account at archevi.ca
- Create your family (tenant) workspace
- Invite family members with appropriate roles
- Upload documents via the web interface, email forwarding, or bulk upload
- Start asking questions using the AI chat
What types of documents can I upload?
Archevi supports:
- PDFs (text extraction + visual search for charts/diagrams)
- Images (JPG, PNG - with OCR for text extraction)
- Office documents (Word, Excel - converted to text)
- Text files (TXT, Markdown)
Documents are automatically categorized into: Insurance, Medical, Financial, Legal, Recipes, Family History, Education, and more.
How are my documents stored and secured?
- Documents are encrypted at rest using AES-256
- Each family has isolated storage (multi-tenant architecture)
- Row-level security ensures you only see your family's data
- All API access requires authentication via JWT tokens
- Optional 2FA for additional account security
What AI models does Archevi use?
Archevi uses a hybrid AI approach:
- Cohere Embed v4: Document embeddings for semantic search
- Groq Llama 3.3 70B: Fast, intelligent chat responses
- Cohere Command R+: Fallback for complex queries
- Claude: Entity extraction and knowledge question generation
Knowledge Refinement System
What is the Knowledge Refinement System?
When Archevi extracts information from documents, it sometimes finds ambiguous entities. For example:
- Is "Rob" the same person as "Robert Hudson"?
- Does "the lake house" refer to your vacation property?
The Knowledge System asks you to confirm these connections, improving future search accuracy.
How do I answer knowledge questions?
The Knowledge page shows AI-generated questions:
- Swipe right (or click checkmark): Confirm the hypothesis is correct
- Swipe left (or click X): Deny - they're different things
- Swipe up (or click skip): Unsure - skip for now
How do I check knowledge system status?
The Knowledge dashboard shows:
- Pending questions awaiting your confirmation
- Questions answered today
- Total confirmed facts
- Monthly activity
How does AI generate knowledge questions?
When documents are processed:
- AI extracts entities (people, places, events, things)
- Compares entities across documents
- Identifies potential matches (same name, nickname, etc.)
- Generates questions for human confirmation
How do confirmed facts improve search?
Once you confirm that "Rob" = "Robert Hudson":
- Searching for "Rob" finds documents about "Robert Hudson"
- The AI knows they're the same person
- Biographies and timelines merge their information
Sharing & Collaboration
How do I share documents with family members?
Within your family:
- Select a document
- Click "Share"
- Choose family members
- They'll see it in "Shared with Me"
How do I create external sharing links?
For sharing outside your family:
- Select a document
- Click "Create Link"
- Set expiry date (optional)
- Set password protection (optional)
- Copy the secure link
How do I view all shared links?
Go to Settings > Shared Links to see:
- All active links
- Access counts
- Expiry dates
- Revoke links as needed
System Health & Monitoring
How do I check if the API is running?
Use the health check endpoint to verify system status:
Returns:
- System status
- Database connectivity
- AI service availability
How do I monitor database health?
The database stats endpoint returns:
- PostgreSQL connection pool status
- Table sizes and row counts
- pgvector index health
- Recent query performance
How do I check vector database status?
Vector (embedding) stats show:
- Total embeddings stored
- Coverage percentage per tenant
- Index health and performance
- Recent embedding operations
What if embeddings are missing?
If documents aren't returning in searches:
- Check if the document has an embedding (green dot indicator)
- Click "Re-embed" to regenerate
- Wait for processing (usually < 30 seconds)
Technical & Troubleshooting
Why are my searches slow?
Possible causes:
- Large document count (optimize with categories)
- Complex queries (simplify or be more specific)
- Network latency (check connection)
- First query after idle (cold start)
Why won't my document upload?
Check:
- File size (max 50MB per document)
- File type (PDF, images, Office docs, text)
- Available storage quota
- Network connectivity
How do I report a bug?
- Note the error message and steps to reproduce
- Go to Help > Report Issue
- Include screenshots if helpful
- We'll respond within 24 hours
Where can I get more help?
- Documentation: docs.archevi.ca
- Email: [email protected]
- Chat: In-app chat bubble
- Community: Discord server
Timeline & Biography
How do I view family timeline events?
The Timeline shows important dates extracted from documents:
- Document expiry dates
- Event mentions in content
- Manually added events
Filter by date range, event type, or family member.
How do I generate AI biographies?
Select a family member and click "Generate Biography". Choose a style:
- Narrative: Story-like prose
- Formal: Professional tone
- Casual: Conversational style
- Children: Simple language for kids
Can I add calendar feeds?
Yes! Subscribe to expiring documents in your calendar:
- Go to Settings > Calendar
- Copy the iCal feed URL
- Add to Google Calendar, Apple Calendar, etc.
- Get reminders for expiring documents
Last synced from CMS: 2026-01-06 11:35