Deploy on your infrastructure, with your own AI.
Measurement data is fab-confidential — so AutoPDK goes to the data, not the other way round. And the intelligence layer is swappable: our hosted models, your enterprise AI subscription, or an air-gapped local model.
Cloud API
Hosted AutoPDK with everything included. Push measurements over a REST API, stream extraction progress live, pull model cards and reports. AI included and managed.
- Fastest path from data to model card
- Live progress events for your own dashboards
- Per-project isolation
Private deployment
The entire platform — extraction engine, SPICE farm, copilot, database — ships as containers and runs on your hardware, inside your network. Air-gapped if required.
- Measurement data never leaves your site
- One-command bring-up on a workstation or a cluster
- Same API and UI as cloud
Bring your own AI
You choose the intelligence. Point AutoPDK at OpenAI, Claude, your enterprise AI gateway, or a fully local llama.cpp server — and run two together, one driving the extraction while a second independently verifies it. One setting swaps providers; no code changes, no lock-in.
Fab-confidential by default.
Your data stays where you put it
Private and air-gapped deployments never send measurement data, model cards or reports outside your network. The AI endpoint is yours to choose — including a fully local model.
Same product everywhere
Cloud, cluster or workstation run the same containers, the same API and the same UI. Start hosted, move on-prem later — nothing to migrate but data.
Agent-native by design
An MCP server exposes the whole platform as tools — connect your own AI agents, scripts or EDA flows and drive extractions, simulations and reports programmatically.
Tell us where it should run.
Hosted, private or air-gapped — pick your deployment when you request access and we set it up with you.