Intelligence API
Structured memory engine that categorizes, prioritizes, and evolves with your business - all in one powerful API.
Intelligence Primitives
Every capability is exposed as a clean API endpoint. Compose them independently or as a unified intelligence layer.
Persistent, structured memory store for long-term context. Write once, retrieve semantically, persist across sessions.
Vector-based retrieval with similarity scoring and configurable top-k results. Powered by embeddings, no infrastructure required.
Chain-of-thought inference with configurable depth and structured output modes. Returns reasoning steps alongside final answers.
Dynamic context injection, window management, and deduplication. Only relevant memory is injected — reducing token waste.
REST and streaming endpoints with SDK support for Python, TypeScript, and cURL. Full OpenAPI spec included.
Developer Experience
One API key. Three endpoints. Ship intelligent features without managing vector databases, embedding pipelines, or reasoning infrastructure.
// Write a memory entry
const response = await fetch('https://api.bizxengine.com/v1/memory/write', {
method: 'POST',
headers: {
'Authorization': 'Bearer YOUR_API_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify({
namespace: 'user_123',
content: 'User prefers concise responses under 100 words.',
tags: ['preference', 'style'],
ttl: 86400 // 24h expiry
})
});
// { "id": "mem_9xKzA", "stored": true, "tokens": 14 }
// ← response
// Semantic search across stored memories
const results = await fetch('https://api.bizxengine.com/v1/memory/search', {
method: 'POST',
headers: {
'Authorization': 'Bearer YOUR_API_KEY'
},
body: JSON.stringify({
namespace: 'user_123',
query: 'What are this user\'s formatting preferences?',
top_k: 3,
threshold: 0.75
})
});
// { "matches": [{ "id": "mem_9xKzA", "score": 0.92,
// "content": "User prefers concise responses..." }] }
// ← response
// Run reasoning with injected context
const answer = await fetch('https://api.bizxengine.com/v1/reason', {
method: 'POST',
headers: {
'Authorization': 'Bearer YOUR_API_KEY'
},
body: JSON.stringify({
prompt: 'Summarise the user\'s onboarding status.',
namespace: 'user_123',
chain_of_thought: true,
output_schema: { type: 'text' }
})
});
// { "steps": [...], "answer": "User completed step 2..." }
// ← response
Built With BizXEngine
From production copilots to research pipelines — the same API, composed differently.
Give your product a memory layer. Copilots powered by BizXEngine remember user preferences, past decisions, and domain context across every session.
Replace fragmented vector DB setups with a single API. Store documents, query semantically, and inject relevant chunks directly into your LLM context window.
Build agents that maintain continuity across long conversations. Context windowing and deduplication prevent drift and repetition without ballooning token costs.
Index internal docs, support tickets, or product specs. Expose a semantic search interface over your private knowledge — without exposing raw data.
Getting Started
Three API calls. No infrastructure setup. No vector database provisioning.
Add your API key. Point to our base URL. No provisioning, no cluster management.
Push memory entries, documents, or structured context to the engine. We handle embedding and indexing.
Retrieve, reason, and respond with full intelligence. Stream results back to your UI in real time.
Pricing
Start free. Scale as you grow. Every plan includes full API access — per workspace, no feature gating.
Forever free. No credit card.
Start for freeBilled monthly. Cancel anytime.
Get startedBilled monthly. Cancel anytime.
Get startedVolume pricing. Dedicated infra.
Talk to usEverything you need to integrate, extend, and ship with the BizXEngine Intelligence API.
Full endpoint documentation — auth, memory, search, reasoning, streaming, and usage events.
Official Python and TypeScript SDKs with full type safety, retry logic, and streaming helpers.
Namespace design, tagging strategies, TTL configuration, and deduplication threshold guides.
Get in touch
Whether you're exploring the API, hitting scale limits, or need an enterprise contract — we respond within one business day.
Message sent!
We'll reply within one business day.