Why We Win

NetApp stores and protects the data; Generate turns it into intelligence — and the two are integrated so tightly that AI inherits NetApp’s data services for free: snapshots, replication, immutable audit, and instant cloning.

Very few AI platforms can operate against a terabyte-scale enterprise file system — and almost none are engineered around NetApp specifically. Most agentic systems choke when asked to vectorize a large file system full of dispersed, mixed-format data. That is exactly where DIY RAG and agent projects fail: teams spend months assembling their own pipeline, hit the scale-and-format wall, and stall. Generate was born from that failure pattern — the turnkey answer to the problem enterprises kept trying, and failing, to build themselves.

Six ways in, none of them announcement-stage:

No — and this is a headline. Generate connects directly to existing ONTAP NFS and SMB shares. No migration, no ETL, no second copy to secure. Data stays exactly where it is, under existing controls. This removes the single biggest objection to enterprise AI projects before it is even raised.

Because that is exactly where the customer’s valuable data already lives — and where cloud AI tools choke or become uneconomical. Generate is built to index and serve RAG over that volume of data in place. It is the reason a storage-heavy account is the perfect account: the more data on NetApp, the stronger the case.

K3s orchestration, vLLM inference, Milvus vector DB, Strimzi Kafka streaming, PostgreSQL, and Grafana monitoring — all on NetApp ONTAP. Models are open, so the customer is never locked to a single vendor’s frontier model. AgentWatch, Iterate’s LLM governance and cost gateway, is available separately for centralized controls across AI infrastructure.

Three steps, about 20 minutes: pre-check, install, done. The partner runs one installer that handles NVIDIA drivers, CUDA, K3s, vLLM, the container registry, model download, and the Generate application services. The customer then opens a browser and starts building agents. No Kubernetes expertise required.

All sizes run the identical platform on Dell PowerEdge servers, NVIDIA or Intel accelerators, and NetApp AFF storage; expanding only adds redundancy and nodes.

Available now: the Better Together one-pager, the Generate on NetApp ONTAP technical guide, AIPod Mini reference designs, and a demo environment through your partner lab. For deal support, references, and installer access: netappsupport@iterate.ai.