Kestrel Labs Brings AI Code Compliance Checking Into Revit
Kestrel Labs is bringing AI code compliance into Revit, giving architects a way to check BIM models against building codes while they design. Engineering News-Record covered the launch on November 19, 2026. The tool targets a task that still eats up long hours on most projects: confirming that a design satisfies every applicable code requirement.
- Where it runs: inside Autodesk Revit, not as a separate web viewer
- What it checks: BIM models against requirements specific to a jurisdiction
- When it checks: during design, rather than at a late review stage
- What it returns: flagged issues for the design team to resolve
- Pricing and availability: not announced
How Kestrel Labs AI code compliance works in Revit
Kestrel reads the Revit model and compares it with the code rules that apply where the project will be built. When something looks out of line, it flags the issue so the architect can fix it in the model.
The pitch is timing. Code problems that surface at permit review or in a plan checker's comments cost far more to fix than ones caught while walls and egress paths are still being placed. Kestrel has not yet shared which code families, editions or local amendments it covers, or how it decides which jurisdiction applies.
How it compares with other code compliance tools
Kestrel joins a growing group of AI tools aimed at code and QA/QC work. ArchiMind takes a similar in-Revit approach but focuses on one state, as we covered in our look at automated California code compliance in Revit. Archiboost AI frames code checking as part of a broader co-pilot for QA/QC, firm knowledge and coordination.
Firms already using tools that mine past projects, such as Pirros for reusing Revit details and QA/QC, should consider where a code checker fits alongside them rather than as a replacement.
What to check before adopting Kestrel
Code checking is high-stakes, so a pilot should test more than the demo. Before rolling it out, BIM managers should ask:
- Coverage: which model codes, editions and local amendments are included for the places you actually work?
- Traceability: does each flag cite the exact code section, so an architect can verify it rather than trust it?
- Model dependence: how much does accuracy depend on clean parameters, room data and families? Messy models may produce noisy results.
- False negatives: what does the tool not check? Missing an issue is riskier than over-flagging one.
- Liability: the licensed professional remains responsible for compliance, so treat the output as a first pass.
A practical test is to run Kestrel on a recently permitted project and compare its flags with the comments the plan reviewer actually issued.
What it means for AEC teams
For architects, moving code review into the authoring tool could shrink the late-stage scramble before permit submission. The real value will hinge on jurisdictional depth and on whether each flag can be traced back to a citable requirement.
Until Kestrel publishes coverage details and pricing, firms should treat it as a promising tool to pilot on a live project, not a substitute for a code consultant or internal review.
Recent Articles
Learn about the latest architecture software, engineering automation tools, & construction technologies
