The construction project simulator

Know how the AI performs before you procure it.

Blocks drops AI agents into complete simulated construction projects. Every RFI, pay application, and statutory deadline is computed by deterministic code, so the findings it reports are ones you can check, with consequences in dollars and days.

Built for construction AI procurements where a single missed statutory deadline can cost more than the software.

Designed for accountable decisions

Construction-specific scenarios

Statute-cited payment, lien, and retainage mechanics, not generic benchmark tasks.

Comparable vendor findings

Every vendor scored on the identical matrix, by deterministic code, never a model judging a model.

Decision-ready reporting

A forwardable report in construction language, with consequence in dollars and days.

Zero real project data

Every scenario is synthetic and jurisdiction-correct. Nothing from your record is ever required.

Why now

Agents are now touching documents with legal consequences.

Construction platforms consolidated the project record and pointed autonomous agents at it. RFIs, change orders, and pay applications sit on statutory clocks: proper-invoice windows, lien-filing deadlines, holdback release. An agent that mishandles a notice deadline doesn't produce a bad summary. It produces a waived claim.

Mistakes are project-wide

On a consolidated record, one bad automated action or one bad access grant touches the whole project (the budget, the bid leveling, the legal file), not one folder.

You can't test on production

Live project records are confidential and contractually off-limits for benchmarking. They also carry no labels for who was entitled to know what, so a permission violation can't be scored after the fact.

No apples-to-apples comparison

Six vendors means six demos, six decks, and six incompatible claims. Nobody on your team has the time, or the ground truth, to referee that honestly.

How it works

A complete project. Real rules. A knowable right answer.

A complete synthetic project

The team, the schedule, and the full document record, generated deterministically under real jurisdiction rules: Ontario's Construction Act prompt payment and holdback, Texas Chapter 28 prompt payment with lien and retainage mechanics. Statute-cited, independently verified, sign-off documented.

Agents work it like production

Vendors connect over MCP, HTTP, SDK, or CI: the same doors production agents use. Each agent acts as a named team member with a server-enforced access tier. Every read, every action, and every refused attempt is recorded.

Deterministic grading, checkable findings

The correct answer is computed by code from the project record itself, never judged by another model. The report is written in construction language, with each finding's consequence stated in the project's own dollars and days.

A finding, not a promise

Every finding is a statement you can check.

FINDING F-01 · JURISDICTIONAL CONFORMANCE
"The agent applied the 120-day nonresidential lien-filing framework to the residential portion of the work. Texas Property Code §53.052(a) sets a shorter window for residential projects."

Consequence, in the project's own units: filed on the agent's recommended schedule, lien rights on $412,300 of subcontract work would have lapsed.

A language model flattens residential and nonresidential lien deadlines into one number. A statute doesn't. Findings like this are checkable against the statute and the project record: no scorecard you have to take on faith, no methodology you can't inspect.

See the full sample report →

Why Blocks

Built so the answer key exists before the agent does.

Ground truth by construction

The same project specification always replays to the identical record. The approved change-order total on day 180 isn't an opinion. It's a computation.

Graded by code, not by a model

No LLM judges anything in scoring. Same input, same score, on demand. The grading method is also open to review by assurance buyers under NDA.

Permission leakage, measured

Access tiers and company scope are enforced server-side on every fact, so we can count what an agent said that its role could not see. A number nobody else can produce.

Statutory clocks, statute-cited

Payment, lien, retainage, and notice mechanics are computed from versioned jurisdiction profiles researched against primary sources and given a final sign-off, never asserted from a model's memory.

Overfitting closed by design

Graded assessments run on held-out project instances. A different instance is a different world, so an agent fitted to the public examples fails a world it hasn't seen, while an agent that knows the rule passes any of them.

Zero real project data

The record is synthetic and jurisdiction-correct. No NDA over a customer record, no confidential-data question, no waiting on permission from a platform's terms of service.

Who it's for

Two sides of the same procurement.

General contractors

Run a multi-vendor agent bake-off on a synthetic project in your jurisdiction: one matrix, one referee, a scored comparison your risk and legal teams can act on.

Run a bake-off →

AI & agent vendors

Get a diagnostic with the two findings sections procurement asks about, permission leakage and jurisdictional conformance, then wire per-release regression into CI.

Assess your agent →

Engineering teams

Connect over MCP, HTTP, TypeScript or Python SDKs, a CLI, or a GitHub Action, with an ungraded sandbox to verify your integration before anything counts.

Explore the platform →

Your next AI procurement deserves a referee.

See a live bake-off, the sample report, and the environment your vendors would run in, all in thirty minutes.

Book a demo