Why build AI on Scalingo ?
Host AI Apps with Built-In Vector Search
Build AI apps with confidence. We handle the infrastructure, so you can focus on intelligence.
Deploy AI-Powered Apps & MCPs
Run the backend of your AI product — from API endpoints to background workers, embedding pipelines and MCPs — on a fully managed PaaS.
With Vector Storage Built-In
Build semantic search, recommendation engines, and retrieval-augmented generation (RAG) systems using managed database services that support vector similarity queries.
On a Fully Managed Platform
Databases, scaling, monitoring, and backups included. Focus on building AI features — not operating infrastructure.
Choose from our vectorial database options
PostgreSQL + pgvector:
Fully managed vector storage
Store embeddings alongside your application data to power semantic search, recommendations, and RAG applications.
Fast Similarity Search
Approximate vector indexes (e.g. IVFFlat) supported by pgvector for efficient similarity queries at scale.
Scalable Storage
Handle large embedding collections using PostgreSQL's proven storage engine and scaling options.
ACID Compliance
Embeddings benefit from PostgreSQL's transactional guarantees (ACID) for consistency and reliability.
Co-located Data
Keep embeddings and metadata in the same database to simplify architecture and reduce operational complexity.
-- Create vector table
CREATE TABLE embeddings (
id SERIAL PRIMARY KEY,
content TEXT,
embedding vector(1536)
);
-- Create HNSW index
CREATE INDEX ON embeddings
USING hnsw (embedding vector_cosine_ops);
-- Similarity search
SELECT content FROM embeddings
ORDER BY embedding <-> $1
LIMIT 10;
// Create k-NN index
PUT /embeddings
{
"mappings": {
"properties": {
"embedding": {
"type": "knn_vector",
"dimension": 1536
}
}
}
}
// Vector similarity search
POST /embeddings/_search
{
"query": {
"knn": {
"embedding": {
"vector": [0.1, 0.2, ...],
"k": 10
}
}
}
}
OpenSearch: Vector search and analytics
Managed OpenSearch for search, logging, and vector-based retrieval use cases.
k-NN Vector Search
Store and query high-dimensional vectors using nearest neighbor indexes.
Fast Retrieval
Optimized for high-performance search workloads.
Scalable Clusters
Multi-node deployments with high availability.
Managed Operations
Multi-node deployments with high availability.
OpenSearch ML
Embedding models and semantic pipelines.
Extend OpenSearch with machine learning capabilities to generate embeddings and power semantic search workflows directly inside your cluster.
Model Deployment
Register and deploy ONNX embedding models (e.g., from Hugging Face) for inference within OpenSearch.
Ingest Pipelines
Automatically generate embeddings when indexing documents.
Search Pipelines
Embed user queries at search time and execute vector similarity searches.
Semantic Applications
Build RAG systems, AI-powered knowledge bases, agents, MCP tools, and multilingual search experiences.
// Register embedding model
POST /_plugins/_ml/models/_register
{
"name": "sentence-transformer",
"version": "1.0",
"model_format": "ONNX"
}
// Create ingest pipeline
PUT /_ingest/pipeline/embeddings
{
"processors": [{
"text_embedding": {
"model_id": "sentence-transformer",
"field_map": {
"text": "embedding"
}
}
}]
}
Compatible Stack
Compatible with your favorite AI tools
Framework-agnostic infrastructure that works with most AI stacks
Database Layer
PostgreSQL + pgvector
Embeddings storage & vector similarity search
OpenSearch + knn + ML
Semantic and vector search
LLM APIs
OpenAI
GPT-4-class models, embeddings
Anthropic
Claude 3-class models
Mistral AI
European LLM provider
Cohere
Command, Embed models
AI Frameworks
LangChain
Python & JavaScript
LlamaIndex
Data framework for LLMs
Data framework for LLMs
Bring your own stack
Vector-Powered AI Use Cases
Real-world AI applications our customers are shipping
Use Cases
RAG Chatbots
AI assistants powered by your data. Answer questions using your docs and knowledge base
AI Knowledge Bases
RAG-powered internal wikis and FAQs. Company knowledge accessible via AI search
Semantic Search
Find content by meaning, not just keywords. Vector search across documentation and articles
Vector Search
Similarity search for embeddings. Find similar content, products, or recommendations
Document Analysis
Extract insights and retrieve information from documents. Document retrieval systems for enterprise data
AI Assistants
Context-aware agents for your application. MCP-compatible assistants with database context.
Start now using our AI capabilities.
