The complete end-to-end ChunkIQ platform — all ingestion connectors, full pipeline processing, hybrid search, and the management portal — deployed entirely within your own Azure subscription.
ChunkIQ Enterprise ships with all four ingestion connectors pre-configured and ready to activate against your Microsoft 365 tenant.
Enumerates all sites and document libraries. Extracts content from every file format including PDF, Word, Excel, PowerPoint, HTML, and plain text with full provenance metadata.
Indexes Teams channel messages, meeting transcripts, and shared files. Exposes a unified search portal for querying across all ingested Microsoft 365 content.
Traverses all notebooks, sections, and pages. Extracts structured text and preserves section hierarchy as metadata for precise retrieval.
Scans personal and shared OneDrive drives across your organisation. Delta query tokens ensure only new or modified files are re-processed on subsequent runs.
All connectors authenticate via Managed Identity — no passwords or API keys stored in code or configuration. Your data never leaves your Azure subscription.
Ingest → Extract → Chunk & Embed → Index. Each stage runs on Azure Functions inside your subscription and can be triggered on demand or on a schedule.
The connector authenticates to your Microsoft 365 tenant and enumerates files across all four source platforms simultaneously. Files are downloaded to Azure Data Lake Storage Gen2 with content-hash deduplication so unchanged files are never re-processed.
A format-aware dispatcher routes each file to the correct extractor. All extraction runs natively — no external OCR or Document Intelligence services required, keeping costs minimal and the pipeline fully portable within your subscription.
Extracted text is split into semantic chunks using a hybrid chunking strategy that respects paragraph boundaries while staying within the token budget. Each chunk is embedded using Azure OpenAI's text-embedding-3-small model.
Chunks are upserted to Azure AI Search with embedding vectors and all metadata fields. The index supports hybrid BM25 + vector search with semantic re-ranking via Reciprocal Rank Fusion — best-in-class retrieval accuracy for RAG pipelines and enterprise search.
Everything in the individual products, plus enterprise-grade deployment, governance, and control — all inside your Azure perimeter.
Every component — Azure Functions, ADLS Gen2, Azure OpenAI, Azure AI Search, and the Laravel portal — is provisioned inside your Azure subscription via IaC templates.
Your documents, extracted text, embeddings, and search index never leave your Azure environment. Managed Identity authentication throughout — no API keys in code.
The Laravel management portal provides a single view across all four ingestion sources — chunks indexed, last run time, active sources, and pipeline health at a glance.
Content hashing and delta tokens ensure only changed files are re-processed. Large tenants stay fast — run times scale with changes, not total corpus size.
Each pipeline stage runs as an independent Azure Function. Trigger the full end-to-end pipeline on a timer, on demand via HTTP, or from the management portal.
Dedicated onboarding, deployment assistance, and priority support. We help you configure your Microsoft 365 app registration and get the pipeline running in your tenant.
ChunkIQ Enterprise gives your organisation full ownership of the pipeline. All sources, all processing, all search — deployed in your tenant, on your terms.