Construction
Who prices the tender when the deadline is on Friday?
An AI agent prepares a price draft for every tender from your past estimates and turns daily site reports into draft invoices. Your estimator checks every figure before anything is submitted.
- Item 1 · Site set-upYour May bid, item 1
- Item 2 · ExcavationYour unit rate 2026
- Item 3 · Drainage pipe PP 200Subcontractor quote B
- Open: asphalt pricesubcontractor quote missing
Approved by your estimator
A Tuesday evening in the office
- Tender
A tender from simap sits on the desk as a PDF and an NPK file, and two subcontractor prices are still missing.
- Site reports
Next to it are last week's daily site reports. A foreman's time-and-materials slip has no hours on it.
- Post-calculation
Whether a site loses money only shows in the post-calculation, once the job is finished.
- Deadline
The bid has to go out on Friday.
‘Construction today is about half administration.’ Dominik Lüthi, spokesperson of the Swiss Contractors' Association (SBV), Radio SRF 1, ‘Treffpunkt’, 25 September 2026 (quote in German, our translation).
What an agent prepares in your office
Each job gets its own agent. It takes site reports and time-and-materials slips the way they reach your office today.
If your trade software can already do the job, or your data can't carry it, we tell you and stop there.
The agent works between your systems
A chat answers one question for one person. An agent does a whole job: it reads the tender, takes unit rates from past estimates and subcontractor quotes from the mailbox. Your rules come with it as skills and plugins, such as surcharges and your standard reservations.
- Tender
- Trade software
- Mailbox
- Site reports
The quote agent
Works by your rules
skillsplugins
Your estimator checksNothing is submitted before it's approved
Underneath, always
- AccessOnly the folders and systems you allow.
- GuardrailsRules the agent works within.
- LogEvery step is written down.
Who decides, and where does the agent run?
At first the agent only suggests. Nothing is sent, submitted or booked until your estimator or your office approves it. Before the first run we set up who may access what, the guardrails and a log.
Unless you choose otherwise, your agents run in your own environment and we operate them. Microsoft 365 is an important place for agents. It shouldn't be the only one, so we keep a second way open: your own server and open models. If you stop, the agents, the rules and the data stay with you.
The job: preparing tenders
In your Microsoft 365
In Outlook, Teams and Copilot, where your team already works.
For public and internal files
On your own server
Open-source agents on a server you control.
Also for confidential files
As private AI
A model that runs under your firm's name.
For secret files
What you have in hand after the trial run
You start with a trial run on 10 of your quotes. We run an agent, read only, over tenders you have already submitted. In a 90-minute session its draft sits beside the estimate you made at the time. Then we give you a clear yes or no.
The side-by-side
Your estimate and the agent's draft next to each other, item by item.
The list of missing data
What the agent still lacks, such as unit rates for one kind of work.
A note on your software
In writing: what your trade software can already do on its own.
Your decision
If it's a yes, the next step is the AI assessment. If it's a no, we stop there.
Firms with 10 to 250 staff start with the trial run, larger firms with the AI assessment. You decide after each step whether to go on. From the AI assessment on, each step has a fixed price, and we bill no hours.
Questions about agents in construction
Does the agent submit tenders on its own?
No. The agent only makes suggestions, and your estimator approves every bid and every figure.
Do our foremen need a new app?
At first we take site reports and time-and-materials slips the way they reach your office today. Whether anything should change becomes clear after the AI assessment.
What if our software vendor brings this out soon?
Then we tell you. In the AI assessment we check in writing what your software does today and what has been announced. If it already does the job, we make no offer for it.
We're a small firm. Is it worth it?
The trial run shows that on tenders you have already submitted. We tell you plainly, even when the answer is no.
Which tender is on your desk right now?
Bring it to the first call. In 30 minutes we'll work out whether it suits a trial run.
30 minutes with Lukas. You leave with at least one recommendation you can use, whether or not we ever work together.
