Software and AI Agents for Machine Builders
A request for a robot cell comes in. Sales works out which variant fits and which options work together. That selection has to produce the bill of materials, the routing and the quote; once the order is placed, purchasing, warehouse and production need the same status. In special-purpose machine building, this chain rests on a few people who also understand the engineering. Software for special-purpose machinery and project manufacturing therefore has to keep the same details together from the request through to production.
What to check on a real order
- At the request: which variant fits, and which options work together?
- At the quote: do the bill of materials, routing and quote come from the same selection, or does someone copy the details by hand?
- After the order: do sales, purchasing, warehouse, production and accounting work from the same status?
- After delivery: what data do the machines at the customer's site send back, and who may see it?
How it fits together at MAFU-SHERPA
MAFU-SHERPA develops AI-controlled CNC automation for small and medium batch sizes. CERTAINCE introduced Business Central there for sales, purchasing, warehouse, production and accounting. The company's own product logic sits in a CPQ configurator we built inside Business Central, because the standard does not represent special-purpose machine building precisely enough. Sales enters the customer requirements in a structured way; the system derives bills of materials, routings and sales quotes. Hundreds of quotes have already been produced automatically this way.
The data keeps flowing after the order. The DATEV export from Business Central is a fixed automation without AI, because the rule is clear. The machine productivity portal shows each customer only the data of its own machines; more than 100 robots send data into it every day. The order in which these projects came about is told in the digitization of MAFU-SHERPA.
What an AI agent can prepare in machine building
Part of this work follows fixed rules; another part looks different in every case. Variant rules belong in a system that applies them reliably. At MAFU-SHERPA that is the configurator; where your existing programs already apply the rules reliably, we use them. Research, requests and drafts can be taken over by an AI agent when a person checks the result.
- Sales research: read prospective customers' websites for machine park, batch sizes and job ads, explain the fit and draft the outreach.
- Requests: compare an incoming request with the available variants, name missing details and draft the questions.
- Service requests: match a report to the right machine and its history and prepare a reply.
- Reconciliations: compare data between ERP, spreadsheets and project status and report discrepancies.
For MAFU-SHERPA's sales team, CERTAINCE built an AI sales assistant that runs in production. It reads manufacturers' websites, scores the fit from 0 to 100 with reasons and drafts an email for the contact's role. A salesperson reviews the draft and sends it from Outlook.
Where the agent stops
The agent does not set a price or commit to a delivery date. It prepares; a person checks and decides. We set how thorough that check is per task, by what a mistake would cost: a research draft needs a quick look, a quote needs an approval.
What the same way of working looks like for requests with drawings is described on AI agents and software for contract manufacturing.
What we build for this is yours: tested, maintainable code. It usually starts with an opportunity assessment that shows which tasks are suitable. If you would rather go through one case directly, let us discuss a concrete workflow.