Integration Options

SmarterAvatar supports flexible integration models to match your infrastructure and requirements.

Architecture Options

Option A: Full SmarterAvatar Stack

We provide everything—you bring your content.

┌─────────────────────────────────────────────────────────────┐
│                    SmarterAvatar Platform                    │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────────────┐  │
│  │   Gemini    │  │   Whisper/  │  │   React Frontend    │  │
│  │   (LLM+RAG) │  │   Deepgram  │  │   + Admin Panel     │  │
│  └─────────────┘  └─────────────┘  └─────────────────────┘  │
└─────────────────────────────────────────────────────────────┘
                              │
                              ▼
                    ┌─────────────────┐
                    │  HeyGen Avatar  │
                    └─────────────────┘

Best for:

• Rapid deployment
• No existing AI infrastructure
• Minimal integration effort

You provide:

• Knowledge base documents
• Branding assets
• API keys (BYOK)

We provide:

• LLM integration (Gemini)
• RAG pipeline
• Frontend application
• Admin dashboard
• Analytics

---

Option B: Hybrid Architecture

Your backend/LLM + Our frontend.

┌─────────────────────────────────────────────────────────────┐
│                     Your Infrastructure                      │
│  ┌─────────────────────────────────────────────────────┐    │
│  │            Your Backend / AWS Bedrock                │    │
│  │  ┌─────────────┐  ┌─────────────────────────────┐   │    │
│  │  │   Claude/   │  │   Your Knowledge Base       │   │    │
│  │  │   Your LLM  │  │   (Bedrock KB / Custom)     │   │    │
│  │  └─────────────┘  └─────────────────────────────┘   │    │
│  └─────────────────────────────────────────────────────┘    │
└─────────────────────────────────────────────────────────────┘
                              │
                              ▼
┌─────────────────────────────────────────────────────────────┐
│                  SmarterAvatar Frontend                      │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────────────┐  │
│  │   API       │  │   Avatar    │  │   Admin + Analytics │  │
│  │   Gateway   │  │   (HeyGen)  │  │                     │  │
│  └─────────────┘  └─────────────┘  └─────────────────────┘  │
└─────────────────────────────────────────────────────────────┘

Best for:

• Existing AWS/AI infrastructure
• Data sovereignty requirements
• Custom LLM fine-tuning
• Regulatory compliance

You provide:

• LLM/RAG backend (e.g., AWS Bedrock)
• Knowledge base
• API endpoint for chat

We provide:

• Frontend application
• Avatar integration
• STT integration
• Admin dashboard
• Analytics

---

Comparison

Aspect	Full Stack	Hybrid
Setup time	Days	Weeks
Integration effort	Minimal	Moderate
LLM control	Via config	Full
Data location	Our infra	Your infra
Cost visibility	BYOK	Direct
Customization	Standard	Extensive

---

Data Flow Comparison

Full Stack

User → SmarterAvatar → Gemini (RAG) → Response → HeyGen → User

Hybrid

User → SmarterAvatar → Your API → Your LLM → Response → HeyGen → User

---

Choosing the Right Option

Choose Full Stack if:

• ✅ You want fastest time-to-deploy
• ✅ You don't have existing AI infrastructure
• ✅ Standard RAG capabilities are sufficient
• ✅ You're comfortable with BYOK model

Choose Hybrid if:

• ✅ You have existing AWS Bedrock / AI infrastructure
• ✅ You need data to stay in your environment
• ✅ You have custom LLM requirements
• ✅ You have compliance requirements

---

Getting Started

Ready to integrate?

• BYOK Setup Guide →
• API Reference →
• Contact us for Hybrid architecture planning