Blueprints AI Version 2: Generate permit-ready construction docs

What changed

Blueprints AI unveiled a second-generation version of its software that became publicly available this week. The upgraded platform allows users to input a project description or sketch and automatically generate construction documents covering architecture, structural, mechanical, electrical, plumbing, and interior design simultaneously. Users can edit designs within the interface, with changes automatically propagating across all related systems and validated against local building code requirements.

The company was incorporated in 2025 in Santa Monica by co-founder Al Drams, an architect and general contractor who trained the AI models using eight years of approved drawing sets from his own design and construction business. Co-founder Coco Cao, a data scientist from Hong Kong, serves as chief commercial officer and handles go-to-market strategy. The pair met at an AI conference and combined their expertise to establish the company. Blueprints AI earned selection to TechCrunch’s Startup Battlefield 200 cohort and unveiled version 2 at TechCrunch Disrupt in San Francisco this week.

The first-generation software already produced code-compliant construction documents from 2D sketches or PDFs and carries pricing between $3,200 to $8,500 monthly based on the selected plan. Custom enterprise pricing is available for larger organizations. The company may introduce a one-time flat-fee option for single-project use in the future. The startup is already developing a third version that will incorporate structural analysis and electrical panel design capabilities.

Why it matters

Blueprints AI targets a documented workflow inefficiency in architecture. According to Drams, architects allocate approximately 80 percent of their time to drafting tasks and only 20 percent to design work. By automating the repetitive documentation phase, the tool aims to shift that ratio toward 20 percent drafting and 80 percent design, enabling architects to concentrate on creative problem-solving rather than routine documentation.

The second-generation capabilities address a practical challenge in managing complex projects: coordinating dependencies across multiple building systems. When one system changes, interconnected systems typically require corresponding updates. By automating these updates and validating them against building code requirements, the software potentially reduces revision cycles and accelerates permit approval timelines.

The platform targets two customer segments: professional architecture, engineering, and design firms on one side, and developers and real estate owners on the other. For design professionals, it could increase project throughput by handling documentation work at scale. For non-professionals in real estate development, it could provide access to professional-quality construction documents without employing an in-house design team.

An important operational reality: licensed architects must review and affix their seal to all documents before permit submission. The technology automates document preparation but does not replace professional judgment or eliminate licensure requirements. The human review and approval process remains mandatory for all submissions.

What to test

Before committing to Blueprints AI, evaluate these critical factors:

• Code compliance verification: Test whether the software’s code checking covers your specific local jurisdiction requirements. Building codes vary significantly by location and type of project. Verify compliance against actual permit submissions in your area.

• Workflow integration: Determine whether the software imports and exports in formats compatible with your existing tools. Test integration with CAD systems and other design applications your team depends on for daily operations.

• Training data limitations: The AI models were trained on eight years of approved drawings from Drams’ firm. Assess whether outputs reflect building practices outside that firm’s regional and historical context and whether the training data covers your project types.

• Cross-system coordination: Version 2 generates structural, mechanical, electrical, and plumbing documents. Verify that these systems actually coordinate properly and meet engineering standards, not merely that they pass an automated code checker.

• Change propagation reliability: Test the claim that a single design change updates the entire project. Determine whether all updates propagate correctly across building systems or whether some changes fail to cascade properly.

• User experience and training requirements: Evaluate the learning curve for your team. Assess whether input methods such as sketches or descriptions integrate smoothly into your existing workflow or require significant user adaptation.

• Liability and responsibility: Clarify contractual responsibility if the software produces non-compliant documents that pass initial review. Obtain legal guidance on liability allocation before deploying this in production environments.

Common questions

Can you use AI for blueprints?

Yes. Blueprints AI generates construction documents from sketches, descriptions, or PDFs. However, the co-founders emphasize that licensed professionals must review and affix their seal to all documents before permit submission. AI handles generation and code validation, but licensed professionals retain responsibility for final approval.

How much does Blueprints AI cost?

First-generation software ranges from $3,200 to $8,500 per month depending on the plan. Custom pricing for enterprise customers is available. The company is considering a one-time flat-fee option for single-project use, though this is not yet confirmed as available.

Does Blueprints AI replace architects?

No. The software automates preparation of construction documents but does not replace architect judgment or licensure requirements. Licensed professionals must review all documents and provide their professional seal before permit submission. The technology handles repetitive drafting work, freeing architects to focus on design decisions.

The conclusion

Blueprints AI addresses a genuine efficiency bottleneck in architectural practice by automating the documentation work that occupies most of an architect’s schedule. The second-generation software’s ability to generate coordinated drawings across multiple building systems and automatically update them against code requirements represents a meaningful advancement beyond simple drafting aids.

The tool is not a replacement for architectural expertise or professional responsibility. It functions as a preparation layer that professional teams use before human review and submission. Real value depends on whether it genuinely saves time in your specific workflow and whether its code validation matches your local jurisdiction’s actual requirements.

The fact that early customers are already using version 2 suggests real-world testing is underway. Version 3’s planned additions of structural analysis and electrical panel design will indicate whether automated generation can expand into specialized engineering work that traditionally requires deep domain expertise. Those capabilities will define the tool’s long-term impact on the profession.

Related reading

• NVIDIA Vera Rubin Architecture Advances AI Factory Standards
• vLLM Decision 3.0: Five Open Decision Models, 0.8B to 27B
• Google DeepMind’s AlphaProof Nexus Solves Open Math Problems