deckent
In developmentAlphaDeckent
Agent Control & Execution Plane for AI agents. Verhex's main project.
A runtime that decides by policy what AI agents may do, and runs the work inside those limits. The tower and the runway in one place.
AI · Consulting · Integration
Process first, then the system
Verhex builds AI solutions for enterprise processes.
Solutions
Three lines of work, one approach: process first, then the system. Pick a solution to see it in the visual.
- 01
We hand the repetitive work in your processes over to AI: reading documents, classifying, reporting and decision support. We design the workflow first, then connect the model to it.
AI service - 02
We map your production, costing, finance, sales, procurement, HR and CRM processes together with you. We set up the ERP around the process, or bring an existing setup back in line with it.
Consulting service - 03
We build the data flow between your ERP, CRM and internal tools. One system of record, traceable transfers, and connections that replace copying by hand.
Integration service
Projects
Products we build ourselves.
- In developmentBeta
Kâtip
A voice meeting assistant that joins the call, transcribes and translates it live, and records notes, decisions and action items when addressed.
- HR PortalIn developmentBeta
HR Portal
A time and attendance platform installed inside your organisation.
- Open source
Xerify
A shell-first, open-source tool for cross-provider verification.
Notes
What we learn in the field, written down.
Note KVKK compliant AI architecture: where should the data live?
AI projects that handle personal data have three architectural options: mask the data before it reaches the model, use an enterprise cloud hosted in a European region, or run an open model on your own infrastructure. The choice is a trade-off between cost and risk, and which data goes where should be put in writing at the start of the project.
Note RAG or fine-tuning? Making the right call in enterprise AI projects
In most enterprise projects the right answer is RAG. Fine-tuning does not teach a model new knowledge; it teaches behaviour and format. Use RAG when your information changes, when answers must cite a source or when your document volume is large. Add fine-tuning when you need a fixed tone, format or domain language.
Have a project? Let's talk.
Tell us briefly and we will get back to you. The first call is not a pitch: we listen to what you need and say plainly whether it can be solved. If you need an NDA, please let us know before the call.