Choosing the process
Together we pick the process that repeats most often, has a clear definition of "done" and has its data in your systems. We define what the AI does on its own and what it does only after a person approves.
AI for construction companies
Estimates are built in CostX or Buildxact, subcontractor bids arrive as Excel and PDF, tenders are published on the public procurement portal, and completion certificates and invoices are checked by hand before they are posted in SAP Business One. AI breaks the documents down, matches line items to your estimate and answers from the project file with a source, while a person makes the decision. We implement it with access rules and data in the EU.
Built for ISO 27001NIS2 DORA
- Work log and photo record summary
- Public tender document analysis
Estimate in CostX, bids in Excel, tender on the portal, and everything compared by hand
- 1 table bids matched to your CostX or Buildxact estimate
- With a source answers from the specification with a page reference
- A week Procore site diary and Teams photos in one summary
- Before payment Buildxact certificates checked before posting in SAP Business One
A construction company's project manager receives bids from subcontractors and suppliers in different formats every day: one in Excel, another a PDF from a CostX or Buildxact estimate, a third in the body of an email. Line items are named differently, units of measure do not match your estimate, and the comparison is due tomorrow. On site, the foreman fills in the site diary in Procore or Excel and sends photos to Teams that nobody has time to review. At the end of the month, subcontractor completion certificates from Buildxact are checked against the contract and the estimate, and then keyed into SAP Business One. In public tender documents from the procurement portal, the requirements are scattered across hundreds of pages, and the decision whether to bid is made only a few days before the deadline.
Here AI works with documents rather than making decisions: it breaks bids down into a table that follows the line items of your CostX or Buildxact estimate, finds the answer in the technical specification with a page reference, summarises the week's site diary from Procore and the photos from Teams, turns a tender package from the procurement portal into a list of requirements, and checks a Buildxact completion certificate against the contract before it is imported into SAP Business One. We implement it with Microsoft 365 Copilot in project files on SharePoint, or with Claude or ChatGPT through the API, and we integrate with CostX, Buildxact, Procore and SAP Business One through the API, file export or an email flow, depending on your version. Before that, we set up access by project and role, keep documents in a cloud in the EU, and for tenders we offer private models in the MDP CLOUD GPU cloud in Lithuania (EU).
What we implement for construction companies
The 5 solutions we implement most often in this industry. Each one starts with a single process, has a human approval step and is measured.
Comparing subcontractor quotes and estimates
Bids received in Excel, PDF and email text are broken down into a table of line items that follows your estimate exported from CostX or Buildxact, units of measure are unified and missing items are flagged. The project manager gets the comparison in SharePoint with price differences and notes, and makes the choice personally.
We build it with: Claude through the API + CostX or Buildxact estimate export + Excel/SharePoint
Project documents assistant with sources
The assistant answers questions from the project file: technical specifications, design narratives, drawing PDFs with markups, building regulations and the contract. Every answer names the document and page it came from, so the engineer checks it in seconds instead of searching hundreds of PDFs.
We build it with: Copilot Studio agent in SharePoint project files + project file with drawing PDFs
Work log and photo record summary
Foreman entries in the site diary in Procore or Excel and photos from site in a Teams channel are summarised into a weekly report: what was completed against the schedule, what is late, which deviations are visible in the photos, which materials are missing. The project manager reviews the summary and sends it to the client without writing it from scratch.
We build it with: ChatGPT through the API with image analysis + Procore or Excel diary + Teams
Public tender document analysis
Tender conditions, the technical specification and the draft contract from the public procurement portal are broken down into a list of requirements: qualification, deadlines, warranties, mandatory documents, penalties. Each requirement links to a page, and the estimator gets a checklist and a list of line items to start the CostX estimate, then decides personally whether to bid.
We build it with: Private AI in the MDP CLOUD GPU cloud + tender document package from the procurement portal + Claude through the API
Checking acceptance certificates and invoices against the contract
Subcontractor completion certificates from Buildxact or Excel and invoices are recognised and matched against the contract and the estimate: quantities, rates, parts already paid, retention. Discrepancies are flagged before signing, and clean documents are prepared for import into SAP Business One or NetSuite, where the accountant approves the batch.
We build it with: Document recognition + Power Automate + SAP Business One or NetSuite import
How we implement it
From the first conversation to a working solution, with measurement and ongoing support.
Access and data
We put access to files and systems in order, give the AI its own identity with the least privileges it needs, turn on an action log and choose where the model runs: your Microsoft 365, Amazon Bedrock in an EU region or our GPU cloud in Lithuania (EU).
Build and test
We build the solution with Copilot, Claude or ChatGPT, whichever fits the task, connect it to your systems and test it on historical documents and enquiries until accuracy reaches the agreed threshold.
Launch and support
We launch with a person in the loop, review the logs and exceptions and adjust the rules. Once the first process runs reliably, we move on to the next one on the list.
Are your processes different?
These solutions are a starting point for a conversation, not a complete list. Tell us which processes repeat in your company, and in one conversation we will tell you what is worth handing to AI and what is not.
Frequently asked questions
Can AI make mistakes when comparing estimates, and how will we notice?
It can, which is why every matched line item links to the source document and to the row in your CostX or Buildxact estimate, and unclear items are flagged separately. The project manager approves the comparison rather than accepting it blindly. It is a faster check, not a replacement for checking.
Does it work with CostX, Buildxact, Procore and SAP Business One?
Yes, but the integration method depends on the version: we usually take CostX and Buildxact estimates and certificates through an Excel or XML export, Procore through the API or an export, and SAP Business One or NetSuite through an import file or the API. Before proposing a solution, we check what your versions allow and choose the integration method. We have no official partnerships with these systems; we work with their public interfaces.
Project documents are confidential. Will they end up in public models?
No. We work with business editions in which data is not used to train models, and for public tenders and sensitive projects we offer private AI in the MDP CLOUD GPU cloud in Lithuania (EU) or Amazon Bedrock in an EU region. Access is separated by project and role, the same way as in SharePoint.
Related services
Document and invoice processing
Invoices are scanned, checked and posted to accounting without manual entry.
AI for financial services companies
KYC documents after an Onfido or Sumsub check, customer enquiries, complaints and reports to the financial regulator: in a financial firm this is…
AI for healthcare providers
Reception calls, record summaries from Cliniko or CGM and the national e-health system, insurer and health insurance billing documents, internal…
AI by industry
The value of AI depends not on the platform but on which processes in your industry repeat.
Expertise confirmed by certifications
This service is delivered by specialists with valid international certifications. Their expertise is confirmed by independent organisations, not by us.
-
Artificial Intelligence Security Professional
AI systems are deployed with protection against data leakage and model abuse.
-
ISACA Advanced in AI Security Management
The AI security programme and its controls are managed following the ISACA methodology.
-
ISACA Advanced in AI Risk
AI risks are identified and managed like any other IT risk.
Let's discuss an AI solution for your construction company
Tell us which processes repeat most in your company. Within 1 business day we will tell you where it makes sense to start and how long it would take.
We reply within 1 business day