From system data to a decision you can act on, Daedalus connects one auditable, reproducible chain. People define the rules; every result has a traceable source.
Three product lines share one data and process model. Use any one alone, or chain them together.
Takes over the clicking and typing people do between systems. Already live: automated invoicing on China’s national tax platform across multiple legal entities, ERP integration, and DingTalk.
The execution layer →Turns scattered actions into auditable, reproducible processes across supply chain, finance, operations and HR. The rules come from your people’s experience, not from a model deciding on its own.
The orchestration layer →Leads, customers, follow-ups, deals and referral attribution in one place. The home screen is an agent conversation: ask about a customer, ask for a number. Every tool is read-only — writes are always a human click. Clone it and one command gets it running.
The application layer →No forms, no sales call. Three minutes to run it on your own machine — not a single row of data leaves your network.
Ask about a customer, ask for a number. The model decides what to search, read and query — and you see every call it makes. All tools are read-only.
Paste a WeChat conversation into the record page and AI turns it into a follow-up note, tasks and a next step. You review, then save. The original text is kept.
Every write is recorded. You can always find out who changed what, and when.
DeepSeek, OpenAI, a local Ollama — any of them. Configure none and the AI entry points simply disappear.
What isn’t working yet is labelled as not working yet. Every claim can be checked.
Walk us through your process. We’ll tell you which one or two product lines fit — and which steps should not be handed to AI at all. The demo runs on your real data, not slideware.