iiDrak is five independently deployable products on one control plane: lakehouse compute, low-code AI applications, model operations, a federated catalog and zero-trust governance. Take one for a focused use case or run all five as a single system — connectors, identity and policy are shared either way. Any source, any data type, in your cloud or your data centre.
Lakehouse compute and storage, notebooks, SQL, streaming pipelines, CDC sync and built-in BI — the data plane everything else runs on.
Low-code canvas for building complete AI applications — automations, RAG systems and LLM agents — by dragging and wiring nodes.
Import and serve models behind traffic-split endpoints, wrap them in PII and prompt-injection guardrails, and hold the estate to EU AI Act controls — plus scheduling, experiments and the registry.
Every connected system indexed in place, with lineage, a 474-term ontology graph, data quality testing, a data marketplace and an AI Context Center.
Zero-trust access policies and compliance enforcement down to the column — covering the other four products at once.
One credentialed adapter, configured once, reused everywhere — as a pipeline node, an AI Studio tool and an AI Factory sync source. 150+ pre-built.
Nexaris is not all-or-nothing. Pick what your use case needs and add the rest later.
| Use case | Deploy | Skip for now |
|---|---|---|
| Legacy lakehouse takeout e.g. migrating off Cloudera | iiDrak Lakehouse + Data Catalog + Guardiuum | AI Studio, AI Factory |
| GenAI application development chat, RAG or agentic apps | Wa-Sul AI Studio + Data Catalog + Guardiuum | Lakehouse, AI Factory |
| Model deployment & inferencing only operating models trained elsewhere | AI Factory + Guardiuum | Lakehouse, AI Studio, Data Catalog |
| Full integrated platform data + AI + governance, one control plane | All five products | — |
Guardiuum appears in every pattern — it has nothing to enforce policy over on its own, so it is always paired with at least one other product.