AI Integration Settings

Overview

Marmend includes AI-powered features to help you work more efficiently:

• Email Composer - Generate professional email templates with AI assistance
• CV Matching - Analyze candidate CVs against job requirements

By default, these features use Marmend's AI service. However, you can optionally configure your own AI provider if you prefer to use your own API keys.

Default AI Service

When you use Marmend's AI features without any custom configuration:

• Marmend provides the AI service at no additional cost (included in your plan)
• No setup required - it works out of the box
• Uses industry-leading AI models for best results

This is the recommended setup for most users.

Using Your Own AI Provider

If your organization requires using your own AI API keys (for compliance, cost control, or preference), you can configure a custom AI provider.

Supported Providers

Provider	Website	Best For
OpenAI	openai.com	General purpose, GPT-4
Anthropic	anthropic.com	Claude models, safety-focused
Google Gemini	ai.google.dev	Google ecosystem integration
xAI Grok	x.ai	Latest Grok models
Ollama (Local)	ollama.com	Privacy, local deployment
Ollama (Remote)	ollama.com	Self-hosted servers

How to Configure

1. Navigate to Administration > AI Integration in your admin panel
2. Toggle "Use custom AI provider"
3. Select your preferred Provider from the dropdown
4. Enter your API Key from your provider's dashboard
5. Choose the Model you want to use
6. Click Save

Getting API Keys

OpenAI

1. Go to platform.openai.com
2. Sign in or create an account
3. Navigate to API Keys section
4. Create a new secret key
5. Copy the key (starts with sk-)

Anthropic (Claude)

1. Go to console.anthropic.com
2. Sign in or create an account
3. Navigate to API Keys
4. Generate a new key
5. Copy the key (starts with sk-ant-)

Google Gemini

1. Go to ai.google.dev
2. Sign in with your Google account
3. Get an API key from Google AI Studio
4. Copy the key (starts with AIza)

xAI Grok

1. Go to x.ai
2. Sign in and access the API section
3. Generate an API key
4. Copy the key

Recommended Models

Provider	Recommended Model	Notes
OpenAI	gpt-4o	Best balance of quality and speed
OpenAI	gpt-4o-mini	Faster, more economical
Anthropic	claude-sonnet-4-20250514	Excellent for writing
Google	gemini-2.0-flash	Fast and capable
xAI	grok-2	Latest Grok model

Using Ollama (Self-Hosted)

For organizations that need to keep data on-premises:

Local Ollama (on your machine)

1. Install Ollama from ollama.com
2. Run ollama pull llama3.2 to download a model
3. Select "Ollama (Local)" as your provider
4. No API key required

Remote Ollama (on your server)

1. Set up Ollama on your server
2. Select "Ollama (Remote)" as your provider
3. Enter your server URL (e.g., https://ollama.yourcompany.com)
4. If using authentication proxy, enter the auth token

Reverting to Marmend's AI

To switch back to using Marmend's default AI service:

1. Go to Administration > AI Integration
2. Toggle off "Use custom AI provider"
3. Click Save

Your custom configuration will be removed, and AI features will use Marmend's service.

Troubleshooting

"API key is invalid"

• Double-check you copied the entire key
• Ensure there are no extra spaces
• Verify the key is active in your provider's dashboard

"Connection failed"

• Check your internet connection
• Verify the API key has not expired
• Ensure your provider account has sufficient credits

"Model not found"

• Verify the model name is spelled correctly
• Check if your API plan includes access to that model
• Try a different model from the same provider

AI features are slow

• Try a faster model (e.g., gpt-4o-mini instead of gpt-4o)
• Check your provider's status page for outages
• Consider using a closer region if your provider supports it

Cost Considerations

When using your own API keys:

• You are billed directly by your AI provider for usage
• Costs vary by provider and model
• Monitor your usage in your provider's dashboard
• Set up billing alerts to avoid unexpected charges

Estimated Costs (approximate)

Provider	Model	Cost per 1K tokens
OpenAI	gpt-4o	~$0.005 input, ~$0.015 output
OpenAI	gpt-4o-mini	~$0.00015 input, ~$0.0006 output
Anthropic	claude-sonnet	~$0.003 input, ~$0.015 output
Google	gemini-flash	~$0.00001 input, ~$0.00004 output

Prices are approximate and subject to change. Check provider websites for current pricing.

Data Privacy

Using Marmend's AI

• Data is processed according to Marmend's privacy policy
• We use enterprise AI agreements with providers

Using Your Own API Keys

• Data is sent directly to your chosen provider
• Subject to that provider's terms and privacy policy
• You maintain full control over your API usage

Feature Availability

AI features require the Email Templates feature flag to be enabled for your organization. Contact your Marmend administrator if AI features are not available.

See Also

• Email Templates - How to use AI-generated email templates
• Technical Documentation - For developers and system administrators