Skip to main content

Quickstart

Get started with the Tensorheart Memory API in 5 minutes.

1. Create an Account​

Sign up to get your API key:

curl -X POST https://memoryapi.tensorheart.com/v1/signup \
  -H "Content-Type: application/json" \
  -d '{
    "name": "My Company",
    "email": "you@example.com"
  }'

Response:

{
  "success": true,
  "data": {
    "id": "key_abc123",
    "name": "Default Key",
    "key": "mem_live_your_secret_key_here",
    "key_prefix": "mem_live_abc",
    "environment": "live",
    "scopes": ["read", "write"],
    "created_at": "2024-01-15T10:30:00Z",
    "expires_at": null
  }
}
Save Your API Key

The full API key is only shown once. Save it securely!

2. Extract Your First Memories​

The Memory API uses intelligent extraction to create memories from conversations and documents. This ensures high-quality, structured memories.

export API_KEY="mem_live_your_api_key"

curl -X POST https://memoryapi.tensorheart.com/v1/query/extract \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "User: Hi, I prefer dark mode interfaces. I work as a data scientist at TechCorp and use Python daily.",
    "content_type": "conversation"
  }'

Response:

{
  "success": true,
  "data": {
    "memories_created": 3,
    "memory_ids": ["mem_abc123", "mem_def456", "mem_ghi789"],
    "extracted_memories": [
      "User prefers dark mode interfaces",
      "User works as a data scientist at TechCorp",
      "User uses Python daily"
    ]
  },
  "meta": {
    "request_id": "req_abc123",
    "processing_time_ms": 450
  }
}
Why Extraction?

Extraction uses AI to identify and structure important information, creating higher-quality memories than manual entry. It also handles deduplication and conflict resolution automatically.

3. Query Your Memories​

curl -X POST https://memoryapi.tensorheart.com/v1/query \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "context": "What are the users UI preferences?",
    "max_memories": 5,
    "search_depth": 0.5
  }'
Control Search Thoroughness

Use search_depth (0-1) to balance speed and accuracy. Higher values search more comprehensively but cost more. Default is 0.5.

Response:

{
  "success": true,
  "data": {
    "memories": [
      {
        "id": "mem_xyz789",
        "text": "User prefers dark mode interfaces",
        "relevance_score": 0.92,
        "metadata": {"category": "preference"}
      }
    ],
    "query_context": "What are the users UI preferences?",
    "total_memories_searched": 10
  },
  "meta": {
    "request_id": "req_def456",
    "processing_time_ms": 120
  }
}

4. Get an AI Answer​

curl -X POST https://memoryapi.tensorheart.com/v1/query/answer \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "context": "What theme should I use for the users dashboard?"
  }'

Response:

{
  "success": true,
  "data": {
    "answer": "Based on the user's preferences, you should use a dark theme for the dashboard. The user has indicated they prefer dark mode interfaces.",
    "memories_used": [
      {
        "id": "mem_xyz789",
        "text": "User prefers dark mode interfaces",
        "relevance_score": 0.92,
        "metadata": {"category": "preference"}
      }
    ],
    "confidence": null,
    "model": "gpt-4o-mini"
  },
  "meta": {
    "request_id": "req_ghi789",
    "processing_time_ms": 850
  }
}

Next Steps​