| Overview |
Scalable AI compute platform built on Ray for deploying and fine-tuning large language models in production. |
Intercom Fin is an AI customer service agent built into the Intercom platform that resolves customer queries by learning from your help center and past conversations. It provides instant, accurate answers and seamlessly hands off to human agents. |
| Pricing |
Pay-per-use ($$-$$$$) |
Paid ($29-99/mo + usage) |
| Key Features |
- Ray-based
- Auto-scaling
- Fine-tuning
- Managed endpoints
- Multi-model
- GPU clusters
|
- AI resolution
- help center learning
- conversation history
- human handoff
- multilingual
- custom answers
- analytics
|
| Pros |
- Built on Ray
- Excellent scaling
- Production-grade
- Fine-tuning support
|
- Deep Intercom integration
- Learns from existing content
- Good handoff to humans
- Transparent pricing per resolution
|
| Cons |
- Complex setup
- Higher learning curve
- Enterprise-focused pricing
|
- Requires Intercom subscription
- Cost per resolution adds up
- Limited to Intercom ecosystem
|
