Engineering and construction
The tenders you never get to bid on.
An AI model, trained on your own past bids and running on a server in your own office, drafts the first answer to every tender question. Your estimators and engineers stay on the work that wins.
Tendering is how the work comes in, and how many tenders you can answer decides how much of it you get a shot at. Most of a bid week is not thinking. It is finding the closest answer you already wrote, collecting the same certificates again, and checking that section 4 still agrees with section 11. The AI model does that first pass and hands your people a draft to argue with.
It drafts on your own hardware. Your team reviews, prices and signs, and no tender document or past bid leaves the office.
- Industry
- Engineering and construction
- Where it runs
- A server in your own office
- Working for you in
- About 8 to 12 weeks
- Investment
- About a junior’s salary, for a year
The challenge
Your margin on a job starts before you win it. Every tender costs a week of the people you can least spare, and most of that week is work they have done before:
- Read the tender pack and pull out the requirements, the deadlines and the evaluation criteria
- Split it into questions and send each one to the right person
- Find the closest answer in a past bid and rewrite it
- Collect the same certificates, insurances, policies, case studies and CVs again
- Check that the answers still agree with each other
- Fill in the portal forms
Answer twice the tenders and you do twice of all of it. That is the real ceiling. It is not how many jobs you could deliver, it is how many you can answer for in the time the client gives you.
Tenders you pass on
A no-bid decided on who is free that week, rather than on whether the job suits you. The ones you skip are not deferred. They are gone.
Delivery pays for it
The person who can answer a technical question is usually the person running a job. Every bid week takes them off it, and answers the same question they answered last month.
Old answers, new errors
An expired certificate. A methodology that names last year’s subcontractor. Section 4 promising what section 11 prices differently. The evaluator finds it, not you.
Our approach
We do not sell you a bid-writing product and we do not put a chatbot on the desktop for people to remember to open. We train an AI model on your own submissions and put it inside the bid you are already running.
- 1We go through a real bid with you and find the steps where the same thinking repeats: the reading, the sorting, the finding, the re-collecting, the cross-checking.
- 2The AI model learns from your own submissions: your past bids, your methodologies, your rates, your project references, the way your firm answers a question. Nobody else has this one.
- 3We connect it to where your documents already live, so nobody is asked to file anything anywhere new. The draft lands where your people already look for it.
- 4Your people review, price and sign off. Every correction goes back into the training, so the next first draft starts closer than the last.
There is no code in any of this, so there is nothing to re-code. No business analyst writing a spec every time something changes, no developer building it, no testing, no UAT, no release, and none of it again six months later. When the work changes we retrain the AI model on the new examples. It is also why a tender in a shape nobody has seen before gets a sensible attempt instead of an error, where a rule set would simply stop.
We train the AI model the way you would train a junior, by showing it the work. The difference is that you train it once, it does not resign, and when the business changes you retrain it rather than hire and start again. What it learns stays with you: your firm’s own way of bidding, held in an AI model you own.
What it handles
Reads the tender pack
Scope, specification, conditions, deadlines and the evaluation criteria, taken out of the pack and the addenda and set out as what the client is actually asking for.
Builds the requirements checklist
Every question, every mandatory attachment, every word limit and page limit, against who owns it and when it is due.
Drafts the first answers
Written from your own past bids, with the source shown on each answer, so a reviewer can see which submission it came from and judge whether that was the good one.
Assembles the evidence pack
Certificates, insurances, licences, policies, project references and CVs: the current version of each, collected and placed where the schedule asks for it.
Checks the answers agree
Before it goes in: the same figures, the same programme, the same subcontractors and the same method everywhere they appear in the submission.
Fills in the standard forms
The parts of the portal and the schedules that are the same every time, filled from what is already in the bid rather than typed again.
What stays with your people
The bid or no-bid decision. The price. The risk position and the contract exceptions. The final sign-off. The AI model flags the calls it is not sure about and waits rather than guessing. You set, task by task, what runs straight through and what always stops for a person, and that line is agreed in the scoping.
It runs in your office, so the question of what leaves it never arises
For a contractor the problem with a cloud AI service is not the price. It is that your rates, your margins and your method are the things you compete on, and putting them through somebody else’s system is a question you then have to answer. This one runs on your own hardware, so there is nothing to answer.
- No tender document or past bid leaves for processing. No upload, no outside service, nobody else holding or learning from your submissions.
- No new third party sees your pricing. Not a policy promise: there is simply nothing transmitted.
- No bill that grows with your volume. The AI model is yours and it runs on your server, so bidding more does not cost more.
- The bid and its trail stay inside your own systems, which is where your record keeping already sits.
The results
What the whole project costs, about a year of a junior’s pay, and it pays that back in a few months · indicative
The AI model absorbs exponentially more of the repeated work inside a bid, so the number of tenders you answer is not set by who happens to be free that week
From the scoping call to the AI model doing the work. What moves the date is how easily your systems connect, not the training and tuning of the AI model · indicative
What that buys is not a smaller team. It is the same people, off the parts of a bid they have written before, and a tender list set by which jobs you want rather than by who is free to answer. The figures above are indicative; on the scoping call we work them out again with your own numbers.
Common questions
Does our pricing leave the building?
No. The AI model runs on a server in your own office. Your rates, your margins and your past bids stay on it. There is no upload, no outside service holding them, and nothing for you to disclose as a new third party.
Who decides whether we bid?
You do. The AI model reads the pack and tells you what is in it, including the things that usually decide a no-bid: the conditions, the insurance levels, the programme, the mandatory experience. The decision is yours, and so is the price.
What if our past bids are messy, scanned or in different formats?
That is the normal starting point, and it is why the AI model is trained on your own submissions rather than configured against a template. It learns the variation. Where a document is genuinely unreadable, an old scan or a bad photocopy, it says so instead of guessing.
What changes for our bid team and estimators?
They stop writing the answer they wrote last month and keep the parts that need them: the method, the price, the risk. This is built so a firm can answer more tenders without adding people at the same rate. It is not a redundancy tool, and it could not be: the corrections your people make are the only thing that keeps it improving.
How long does it take?
About 8 to 12 weeks to a working system, with the scoping before it. What moves that is not the bid work. It is how your systems are set up and how much has to be built to reach your documents.
Is your first stretch the same as this?
A scoping call is a conversation about which of your tasks are worth taking on, and what they are costing you now. It is free, and there is no obligation.
Prefer to talk? Call +61 475 043 070