Questions plan review teams ask about Mason AI.

Mason reviews residential plan sets for private plan reviewers and performs an AI-powered first pass. It drafts likely code comments, prioritizes failed checks, links each check to the relevant code reference, and gives the reviewer a faster starting point for the final comment letter.

Mason AI is built for private plan reviewers, third-party permitters, and private providers alike so they can accelerate plan review.

A reviewer uploads the plan set, and Mason reads the drawings with computer vision, runs code checks, and presents the likely failures first. Passed and non-applicable checks remain available so the reviewer can confirm what Mason checked instead of treating the AI plan review output as a black box.

Mason reviews produce a pre-drafted comment letter with an average of sixteen AI-drafted comments per letter. Reviewers can accept, edit, reorder, delete, or add comments before exporting the final letter.

Yes. When Mason checks a code issue, it links to the relevant code reference and includes embedded code books inside the product, so reviewers can verify the cited section without leaving the review screen. Mason checks stay tied to source material instead of floating as unsupported AI suggestions.

For Residential projects, Mason checks California Residential Code architectural requirements, California Plumbing Code, California Electrical Code, California Wildland Urban Interface Code, and the California Energy Code, including Title 24 energy standards.

Mason AI's current residential scope does not include California Residential Code structural checks, California Mechanical Code, California Building Code, CalGreen, or solar permits. The product is explicit about these limits so reviewers know where human review still carries the load.

Local building code amendments require reviewer verification, and local preferences or practices require reviewer judgment. Mason checks can support the review process, but Mason AI does not pretend those jurisdiction-specific decisions are fully automated.

Mason AI reports 96% expert agreement with its code checks. Mason reads walls, doors, windows, and fixtures with a floorplan computer-vision model that reaches 97% accuracy before the reviewer touches the file.

Mason AI invests heavily in evaluation: half of engineering time goes to tests and synthetic edge cases. The product is built to catch regressions that matter in AI plan review, including small geometry changes that can affect code decisions.

Yes. Mason AI shows what it sees, links checks to code sections, and keeps passed, failed, and non-applicable statuses visible. Mason reviews are meant to be inspected by a professional, so reviewers can cross-check the AI reasoning against the code and the drawing before adding anything to the letter.

Working with Mason reduces blank-page work by drafting the first-pass comment letter and surfacing likely failed checks first. The site reports 40% faster architectural review of Residential projects.

Yes. Working with Mason gives entry-level reviewers another round of quality assurance by surfacing likely fire and life safety issues, attaching citations, and making the full set of checks visible for review.

Working with Mason is designed to make response letters more consistent in style, tone, and code citations. Reviewers still control the final wording, but they start from a shared AI-drafted structure instead of separate blank documents.

Yes. Mason AI offers a free trial for trying the tool, including autocomplete comments, up to four plans per month, embedded code book access, and email support.

The enterprise plan is custom for teams and organizations. The enterprise plan includes autocomplete comments, CRC architectural, CEC, CPC, CWUIC, and Energy Code checks, bespoke integrations, 24/7 priority support, and a shared Slack or Teams channel.

Mason AI supports Word export, and enterprise plans can support exports into a jurisdiction's preferred format.