Freeda
AI for automating construction plan reviews to detect errors in 48 hours
Website: https://www.freeda.so
Cover Block
PUBLIC
| Attribute | Detail |
|---|---|
| Name | Freeda |
| Tagline | AI for automating construction plan reviews to detect errors in 48 hours |
| Headquarters | Paris, France |
| Founded | 2024 |
| Stage | Seed |
| Business Model | SaaS |
| Industry | Proptech |
| Technology | AI / Machine Learning |
| Geography | Western Europe |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding Label | Seed (total disclosed ~$3,900,000) |
Links
PUBLIC
This section provides direct links to Freeda's primary online presence. The company maintains a website and a LinkedIn company page, which are the standard public channels for a startup at this stage. No other official social media profiles or product repositories were identified in the available research.
- Website: https://www.freeda.ai
- LinkedIn: https://www.linkedin.com/company/freeda-ai
Executive Summary
PUBLIC
Freeda is a venture-scale bet that AI can de-risk the costly, manual process of reviewing construction blueprints, a wedge into a global industry where schedule overruns and compliance fines are endemic [TechFundingNews, 2025]. Founded in late 2024, the Paris-based startup combines computer vision and machine learning with a layer of human expert review to detect errors, regulatory violations, and inconsistencies in plans within a 48-hour window, a significant acceleration over traditional methods [Finsmes, Nov 2025]. The founding team blends urban planning and technical backgrounds, led by Peter Starr, a former planner with degrees from UCL and HEC Paris, who launched the company based on firsthand experience with the inefficiencies of manual review [TechFundingNews, 2025].
Initial traction, sourced from founder statements, points to over ten clients and the review of 10,000 plans in 2025, with an ambitious goal to analyze one million plans by 2026 [TechFundingNews, 2025] [FoundersToday, 2026]. The business model is SaaS, with a €3.4 million ($3.9 million) seed round closed in November 2025 led by Frst with participation from Brick & Mortar Ventures, capital earmarked for team expansion and product development [TechFundingNews, 2025]. Over the next 12-18 months, the key inflection points will be the validation of its partnership with Socotec, the named engineering firm, the conversion of its reported client pipeline into verifiable, named enterprise contracts, and the technical demonstration that its AI can scale across diverse regional building codes without a degradation in accuracy [Batinfo, 2026].
Data Accuracy: YELLOW -- Core funding and founding facts are confirmed by multiple regional outlets; traction and partnership claims rely on single-source, company-sourced reports.
Taxonomy Snapshot
| Axis | Classification |
|---|---|
| Stage | Seed |
| Business Model | SaaS |
| Industry / Vertical | Proptech |
| Technology Type | AI / Machine Learning |
| Geography | Western Europe |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding | Seed (total disclosed ~$3,900,000) |
Company Overview
PUBLIC
Freeda emerged in late 2024 as a Paris-based attempt to apply machine learning to a specific, costly bottleneck in construction: the manual review of architectural and engineering plans. Founded by Peter Starr, an urban planner with a background at AECOM and degrees from UCL and HEC Paris, the company was conceived to reduce the weeks-long process of checking blueprints for errors down to a 48-hour automated review [HEC Stories, 2026] [TechFundingNews, 2025]. The founding team, which includes co-founders Augustin Perraud, Charles Desbaux, and Mariano Rodriguez, brought together backgrounds in planning, business, and technology, establishing its operational base in France [LinkedIn, 2026].
Beyond its French headquarters, the company has registered a legal entity in the United Kingdom, FREEDA.AI LIMITED (company number 15154551), indicating early plans for cross-Channel operations [Companies House, 2026]. Its primary milestone to date is a €3.4 million (approximately $3.9 million) seed financing round closed in November 2025, led by the French venture firm Frst with participation from U.S.-based Brick & Mortar Ventures [TechFundingNews, 2025] [Finsmes, Nov 2025]. The capital was earmarked for team expansion and product development.
Early commercial activity, as reported by the company, includes serving over ten clients and reviewing 10,000 construction plans during 2025 [TechFundingNews, 2025]. A partnership with Socotec, a major international testing, inspection, and certification group, was announced for joint initiatives throughout 2026, though the specific commercial terms and scope are not detailed in public filings [Batinfo, 2026].
Data Accuracy: YELLOW -- Key facts (founding year, funding round, HQ) are corroborated by multiple regional press outlets. Traction metrics and partnership details are sourced from single reports or company announcements without independent verification.
Product and Technology
MIXED
Freeda's core proposition is to automate a historically manual and time-intensive process: the review of architectural and engineering blueprints for compliance and quality. The company's platform is designed to ingest construction plans and return a detailed analysis of potential errors within 48 hours, a timeline it contrasts with the weeks a traditional manual review can require [TechFundingNews, 2025]. The service is positioned not as a replacement for human experts, but as a force multiplier that combines artificial intelligence with a curated panel of architects, engineers, and specialists in areas like fire safety and accessibility [Finsmes, Nov 2025].
The underlying technology stack is described as a combination of computer vision and machine learning, which allows the system to parse complex drawings and understand elements beyond simple text recognition [Tech.eu, Nov 2025]. According to founder statements, the AI is trained to comprehend plan topology, spatial relationships, measurements, and the nuances of regional building regulations [TechFundingNews, 2025]. This enables the detection of a range of issues, including hidden errors, regulatory violations, and internal inconsistencies that might be missed in a manual pass. The company has publicly announced a partnership with Socotec, a major testing, inspection, and certification group, for joint initiatives focused on using AI to address compliance and quality challenges in construction plans [Batinfo, 2026]. This suggests an early focus on integrating with established industry workflows.
Publicly available traction metrics center on volume processed, not specific product performance benchmarks. The company reports having reviewed over 10,000 plans for more than ten clients in 2025 [TechFundingNews, 2025]. It has also stated an ambition to analyze one million plans by 2026 [FoundersToday, 2026] [EUIS, 2026]. These figures indicate an initial market acceptance of the automated review concept, though they do not detail error detection accuracy rates or client retention.
Data Accuracy: YELLOW -- Product claims are sourced from founder quotes in press releases and regional tech publications. The Socotec partnership is noted in an industry trade publication. Core traction metrics (10,000 plans) are not independently verified.
Market Research
PUBLIC The market for automated construction plan review is emerging from a confluence of chronic industry inefficiency and new regulatory pressure, creating a clear opening for software that can de-risk projects before ground is broken.
Quantifying the total addressable market for this specific service is difficult, as no third-party analyst report sizing the global construction plan review software market was cited in the available research. A comparable market size can be inferred from adjacent sectors. The global construction software market is projected to reach $22.5 billion by 2028, growing at a compound annual rate of 8.5% [Grand View Research, 2023]. Within this, the broader architectural, engineering, and construction (AEC) software segment, which includes design, modeling, and project management tools, is a multi-billion dollar category. Freeda's initial wedge targets a sub-segment of this market: the manual, expert-driven review of blueprints for compliance and errors, a process the company claims can cost thousands per project and take weeks [TechFundingNews, 2025].
Several demand drivers underpin the opportunity. The primary tailwind is the persistent and costly nature of construction errors. Rework due to design and coordination mistakes can consume 5% to 10% of total project costs in the industry [McKinsey, 2017], a figure that creates a strong economic incentive for tools that catch issues earlier. A second driver is increasing regulatory complexity, particularly around sustainability (e.g., embodied carbon calculations), accessibility, and fire safety, which expands the surface area for compliance checks. The company's partnership with Socotec, a major testing, inspection, and certification firm, is positioned to address these compliance challenges directly [Batinfo, 2026]. Finally, a generational shift toward digital construction workflows, accelerated by Building Information Modeling (BIM) adoption, makes the industry more receptive to AI-powered analysis layered on top of existing digital plans.
Key adjacent and substitute markets define the competitive context. The most direct substitute is the incumbent process of manual review by in-house or third-party experts, which Freeda aims to augment rather than fully replace. Adjacent markets include broader construction project management software (e.g., Procore, Autodesk Construction Cloud), BIM coordination platforms, and code compliance software. Freeda's differentiation rests on a deep, plan-specific focus rather than horizontal project management. Regulatory forces are a double-edged sword; while new building codes create demand for automated checks, they also introduce geographic fragmentation, requiring the AI to be trained on region-specific regulations,a challenge the company acknowledges by highlighting its understanding of "regional regulatory details" [TechFundingNews, 2025].
Given the absence of a cited, dedicated TAM, the following table presents analogous market sizing from public reports to frame the broader opportunity.
| Market Segment | Size (Projected) | Growth Rate (CAGR) | Source |
|---|---|---|---|
| Global Construction Software Market | $22.5B by 2028 | 8.5% | [Grand View Research, 2023] |
| Cost of Rework in Construction | 5-10% of project cost | (Industry benchmark) | [McKinsey, 2017] |
This framing suggests the serviceable market is a meaningful slice of a large and growing software category. The economic driver,avoiding costly rework,is well-established, but the specific willingness to pay for a standalone AI review tool, separate from bundled project management suites, remains unproven at scale.
Data Accuracy: YELLOW -- Market sizing is drawn from analogous, dated third-party reports; specific TAM for plan review is not confirmed. Demand drivers are supported by general industry analysis.
Competitive Landscape
MIXED Freeda enters a market defined by manual processes, positioning its AI as a direct replacement for the slow, expensive, and error-prone human review of construction documents.
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Freeda | AI-powered automated review of construction blueprints for errors and compliance. | Seed, ~$3.9M (2025) | Combines computer vision, ML, and human expert review to deliver results in 48 hours. | [TechFundingNews, 2025]; [Finsmes, Nov 2025] |
| Socotec (Partner) | Global testing, inspection, and certification (TIC) company. | Publicly traded, large enterprise. | Offers traditional manual review services; partnered with Freeda for AI initiatives in 2026. | [Batinfo, 2026] |
The competitive map is not dominated by pure-play software startups. The primary alternative is the entrenched incumbent service model: large engineering and TIC firms like Socotec, Bureau Veritas, and AECOM, which employ teams of architects and engineers to manually check plans. This is a high-cost, time-intensive service, often taking weeks, which Freeda explicitly targets [TechFundingNews, 2025]. Adjacent substitutes include general-purpose design validation software within BIM (Building Information Modeling) platforms like Autodesk Revit, which can flag clashes and basic rule violations but are not optimized for deep, regulation-specific compliance checking across disparate drawing sets and regional codes.
Freeda's current defensible edge is its specific technical focus and early commercial traction within a niche. The platform's stated ability to understand plan topology, measurements, and regional regulations suggests a data moat is being built from the 10,000 plans reviewed in 2025 [TechFundingNews, 2025]. This corpus of annotated construction documents, validated by human experts, is proprietary training fuel. The partnership with Socotec is a significant, if early, distribution advantage, providing a channel into a large, established client base and a source of domain expertise [Batinfo, 2026]. This edge is perishable, however, if a well-funded competitor with superior engineering talent or an existing large dataset decides to prioritize this use case.
The exposure is twofold. First, Freeda is vulnerable to competition from the very incumbents it seeks to disrupt, should they develop or acquire similar AI capabilities. A firm like Autodesk, with deep BIM integration and vast industry reach, could theoretically bundle automated code checking, rendering a point solution less attractive. Second, the company's reliance on a hybrid AI-human model, while a current strength, creates a scalability and margin challenge. The "human expert review" component cited in its process [Finsmes, Nov 2025] must be efficiently scaled or largely automated for the unit economics to support a high-growth SaaS model, a technical hurdle that remains unproven at scale.
The most plausible 18-month scenario hinges on execution speed and partnership depth. If Freeda successfully ingests its targeted one million plans by 2026 [FoundersToday, 2026], refines its models to reduce human-in-the-loop dependency, and expands the Socotec partnership beyond pilot projects into a reseller agreement, it becomes the de facto software layer for automated plan review in Europe. In this scenario, slower-moving incumbents become the losers, ceding margin and client relationships. Conversely, if product development lags or the AI accuracy fails to meet enterprise reliability standards, Freeda loses. The winner in that case would be the incumbent service providers, who would retain their fee-for-service model while incrementally adopting cheaper, off-the-shelf AI tools from larger software vendors.
Data Accuracy: YELLOW -- Competitive analysis is based on the subject's stated positioning and a single confirmed partnership; the broader competitor set is inferred from the market structure as no other named AI competitors were identified in available sources.
Opportunity
PUBLIC
If Freeda can systematically reduce the multi-billion-dollar cost of construction errors, the company could become the default quality control layer for the global building industry.
The headline opportunity is to establish a de facto standard for automated building code and design compliance. The construction industry is notoriously slow to adopt new software, but the financial and regulatory pressures are mounting. Freeda's wedge is not just speed, but comprehensiveness,its system aims to understand plan topology, measurements, and regional regulations [TechFundingNews, 2025]. Should it achieve high accuracy across a critical mass of building codes, it could become a non-negotiable insurance policy for developers and contractors before breaking ground. The seed backing from Brick & Mortar Ventures, a specialist proptech fund, suggests investors see a path to this infrastructural role.
Growth is not monolithic; several distinct scenarios could drive scale. The company's early traction with over ten clients and 10,000 plans reviewed in 2025 [TechFundingNews, 2025] provides a small but tangible launchpad.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Regulatory Mandate | Local governments adopt automated plan-checking to accelerate permitting. Freeda's AI becomes the approved or recommended tool. | A major municipality or national agency pilots the technology, leading to a formal procurement. | The cited partnership with Socotec, a major French construction consultancy, to address compliance challenges is a direct step toward influencing regulatory bodies [Batinfo, 2026]. |
| Enterprise Land-and-Expand | A top-10 global engineering or construction firm signs an enterprise-wide deal after a successful pilot on one major project. | A public case study from a named, tier-one customer demonstrates significant time and cost savings. | The founding team's background includes urban planning and civil engineering at firms like AECOM [HEC Stories, 2026], providing relevant industry credibility for enterprise sales. |
| API Platform | Freeda's core verification engine becomes an embedded service within larger architecture and BIM software suites. | A strategic partnership or integration with a major CAD/BIM provider like Autodesk. | The product's stated technical approach,combining computer vision and machine learning to read complex drawings [Finsmes, Nov 2025],is inherently API-friendly, suggesting a platform pivot is architecturally feasible. |
Compounding for Freeda would manifest as a data and accuracy flywheel. Every plan reviewed adds to a proprietary dataset of construction drawings, annotations, and error patterns. This dataset, unique in its regulatory specificity, would train more accurate models, which in turn attract more customers who generate more data. The company's stated ambition to analyze one million plans by 2026 [FoundersToday, 2026] is a quantitative target for this flywheel's initial spin. Early signs of this dynamic are not yet publicly visible in the form of published accuracy improvements, but the model's design implies the intent.
The size of the win can be framed by the cost of the problem, not by a direct public comparable. Rework due to errors and omissions can consume 5-10% of total project costs in construction. For a global industry worth over $10 trillion annually, even capturing a fraction of that waste represents a substantial opportunity. A plausible scenario outcome could see Freeda achieving a valuation comparable to other vertical SaaS companies that digitize critical, high-stakes workflows in fragmented industries. While no direct public comp exists, the seed valuation of $14.5 million [Perplexity Sonar Pro Brief] provides a baseline from which successful execution of any major scenario could drive an order-of-magnitude increase.
Data Accuracy: YELLOW -- Growth scenarios and market framing rely on company-stated ambitions and early partnership announcements; the core problem size (cost of rework) is a well-documented industry figure.
Sources
PUBLIC
[Batinfo, 2026] Partnership with Socotec on AI to rework reading of architectural plans and address compliance and quality challenges; joint initiatives throughout 2026 | https://www.batinfo.com/actualite/socotec-freeda-ia-lecture-plans-architecture-conformite-qualite_197541
[Companies House, 2026] UK-registered entity: FREEDA.AI LIMITED (company number 15154551) | https://find-and-update.company-information.service.gov.uk/company/15154551
[EUIS, 2026] Freeda raises €3.4M to transform construction plan reviews with AI | https://euis.eu/freeda-raises-eur34m-to-transform-construction-plan-reviews-with-ai/
[Finsmes, Nov 2025] Freeda Raises €3.4M in Funding | https://www.finsmes.com/2025/11/freeda-raises-e3-4m-in-funding.html
[FoundersToday, 2026] Freeda raises €3.4M to bring AI-Powered Accuracy to Construction Plan Verification | https://www.founderstoday.news/freeda-raises-over-3m-in-investment/
[Grand View Research, 2023] Global Construction Software Market | https://www.grandviewresearch.com/industry-analysis/construction-software-market-report
[HEC Stories, 2026] Peter Starr (M.20): Building a world without mistakes with Freeda | https://hecstories.fr/en/peter-starr-m-20-batir-un-monde-sans-erreurs-avec-freeda/
[LinkedIn, 2026] Augustin Perraud - Freeda | https://www.linkedin.com/in/augustinperraud/
[LinkedIn, 2026] Mariano Rodríguez - Freeda | https://www.linkedin.com/in/rdguez-mariano/
[McKinsey, 2017] Cost of Rework in Construction | https://www.mckinsey.com/industries/capital-projects-and-infrastructure/our-insights/imagining-constructions-digital-future
[Tech.eu, Nov 2025] Freeda raises €3.4M to transform construction plan reviews with AI | https://tech.eu/2025/11/05/freeda-raises-eur34m-to-transform-construction-plan-reviews-with-ai/
[TechFundingNews, 2025] Freeda bags €3.4M to spot hidden construction errors in 48 hours | https://techfundingnews.com/freeda-raises-3-4m-to-automate-construction-error-detection/
Articles about Freeda
- Freeda's AI Scans 10,000 Blueprints for the Hidden Error — The Paris startup, backed by Frst and Brick & Mortar Ventures, aims to shrink weeks of manual plan review to 48 hours.