RAKE ML

Cover Block

PUBLIC

Attribute	Detail
Company Name	RAKE ML
Tagline	AI platform for predicting and preventing physical asset failure in the built environment.
Headquarters	New York City, United States
Founded	2024
Stage	Pre-Seed
Business Model	SaaS
Industry	Proptech
Technology	AI / Machine Learning
Geography	North America
Growth Profile	Venture Scale
Founding Team	Co-Founders (2)

Links

PUBLIC

• Website: https://www.rakeml.com
• LinkedIn: https://www.linkedin.com/company/rake-ml

Executive Summary

PUBLIC RAKE ML is building a vertically integrated AI platform to predict and prevent physical asset failures in the built environment, a bet that merits attention for its technical approach to a historically opaque and costly problem. Founded in 2024, the company's initial wedge is roof degradation, using what it describes as causal AI to simulate failure scenarios based on fundamental physics, a step beyond standard image recognition [rakeml.com, retrieved 2024]. The founding team pairs a CEO with a military aviation and MBA background with a CTO focused on the technical build, though neither founder's public record includes a prior startup exit in this specific domain [LinkedIn, retrieved 2024]. Capitalization is not publicly disclosed, with no funding rounds or named investors yet visible in primary sources, and the business model is positioned as a SaaS offering for property owners, insurers, and asset managers [PitchBook, retrieved 2024]. The critical watch items for the next 12-18 months are the transition from technical development to commercial validation, specifically the signing of initial named enterprise customers and the securing of institutional capital to scale the platform.

Data Accuracy: YELLOW -- Core product claims are sourced from the company website; founder backgrounds are confirmed via LinkedIn. No independent verification of funding, customers, or technical performance.

Taxonomy Snapshot

Axis	Classification
Stage	Pre-Seed
Business Model	SaaS
Industry / Vertical	Proptech
Technology Type	AI / Machine Learning
Geography	North America
Growth Profile	Venture Scale
Founding Team	Co-Founders (2)

Company Overview

PUBLIC RAKE ML is a 2024-founded startup headquartered in New York City, operating as a privately held entity focused on the built environment [PitchBook, retrieved 2024]. The company's public narrative positions it as a vertically integrated AI platform, though its initial commercial wedge appears to be the prediction of roof degradation and related physical asset failures [rakeml.com, retrieved 2024].

The founding team consists of two co-founders. Rafael Ortega, the CEO, is a U.S. Naval Aviator and a University of Chicago Booth MBA graduate [Perplexity Sonar Pro Brief, retrieved 2024]. His co-founder, Kevin Kabaria, serves as the CTO and is based in New York [Perplexity Sonar Pro Brief, retrieved 2024]. Public milestones are sparse, with the most notable development being a May 2025 LinkedIn post in which the company announced it was expanding its commercial pipeline with what it termed "Risk Signal Solutions" [LinkedIn, May 2025].

Data Accuracy: YELLOW -- Founders and HQ confirmed via multiple public profiles; commercial milestone is a self-published update.

Product and Technology

MIXED RAKE ML positions its core offering as a vertically integrated AI platform designed to convert physical assets into computational intelligence. The company's public materials focus on a specific, high-value application: predicting and preventing roof and building envelope failure. This is framed not as a simple image classifier but as a system that understands the fundamental physics of degradation, translating geospatial imagery and weather data into actionable condition signals [rakeml.com, retrieved 2024].

The platform's utility is articulated across three primary use cases, each tied to a distinct commercial workflow. For transactions, it aims to identify roof and envelope risk before an acquisition closes, supporting pricing adjustments and due diligence [rakeml.com, retrieved 2024]. In insurance, it proposes to generate objective condition evidence to support underwriting, renewals, and claims assessments [rakeml.com, retrieved 2024]. For ongoing asset management, the system is designed to monitor condition changes across a portfolio and translate those signals into repair and replacement timing for capital planning [rakeml.com, retrieved 2024]. A key technical claim is that the platform generates its predictive data by simulating millions of physically-grounded degradation and failure scenarios, a method that suggests a move beyond reliance solely on historical imagery [rakeml.com, retrieved 2024].

• Technical wedge. The differentiation centers on causal AI and physics-based simulation, which the company contrasts with traditional inspections and simple pattern recognition [Perplexity Sonar Pro Brief, retrieved 2024].
• Output granularity. The platform claims to detect specific damage types like hail impact, missing shingles, and algae growth with "claim-ready accuracy" for both residential and commercial properties [rakeml.com, retrieved 2024].
• Commercial signal. A May 2025 LinkedIn post announced an expansion of the commercial pipeline for "Risk Signal Solutions," though no specific customer names or contract values were disclosed [LinkedIn, May 2025].

The technology stack is not detailed publicly. No roadmap for future product surfaces or features has been announced.

Data Accuracy: YELLOW -- Product claims are consistently sourced from the company's own website and a single LinkedIn post. The causal AI and simulation methodology, while described, lacks independent technical validation or detailed case studies.

Market Research

PUBLIC

The market for predictive maintenance in the built environment is moving beyond reactive inspections, driven by the escalating financial risk of asset failure and the increasing availability of high-resolution geospatial data.

A formal TAM, SAM, or SOM for a niche like causal AI for roof degradation is not publicly available from third-party reports. However, the addressable market can be framed by adjacent, well-researched sectors. The global property insurance market, a primary beneficiary of more accurate risk assessment, was valued at approximately $1.7 trillion in direct premiums written in 2023 [Swiss Re Institute, 2024]. The commercial real estate asset management software market, which includes capital planning tools, is projected to reach $15.2 billion by 2028 [MarketsandMarkets, 2023]. These analogous markets indicate the substantial financial flows and operational budgets that a solution like RAKE ML's could intersect with, even if its specific wedge is a fraction of these totals.

Demand is propelled by several converging tailwinds. Climate change is increasing the frequency and severity of weather events like hail and windstorms, directly accelerating roof degradation and insurance claims [National Oceanic and Atmospheric Administration, 2024]. This creates acute pressure on insurers to improve underwriting accuracy and on asset owners to preemptively manage capital expenditures. Simultaneously, the cost of physical inspections remains high and logistically challenging at portfolio scale, creating a clear efficiency gap. The proliferation of satellite and aerial imagery, coupled with advances in computer vision, provides the raw data layer necessary for remote assessment at scale.

Key adjacent and substitute markets highlight both opportunity and competitive pressure. The broader proptech analytics space includes companies offering portfolio-wide energy efficiency monitoring and general facility management platforms. More directly, the insurtech sector is heavily invested in computer vision for claims processing and underwriting, though often focused on post-event damage assessment rather than predictive failure. Traditional substitutes remain manual inspections conducted by engineering firms and roofing contractors, a market defined by its labor intensity and lack of scalability.

Regulatory and macro forces are generally supportive but introduce complexity. Building codes and insurance regulations are evolving, particularly in regions prone to extreme weather, which could mandate more rigorous, data-backed condition assessments. However, the adoption of AI-derived insights in formal underwriting or due diligence processes would require validation from industry bodies and likely a track record of proven accuracy, presenting a significant go-to-market hurdle for any new entrant.

Property Insurance Premiums (2023) | 1700 | $B
CRE Asset Management Software (2028 est.) | 15.2 | $B

The sizing context suggests RAKE ML is targeting a high-value intersection within two massive, established industries. The bet is not on creating a new market, but on capturing a slice of existing spend by offering a superior, data-driven alternative to traditional methods.

Data Accuracy: YELLOW -- Market sizing is drawn from analogous, third-party industry reports; the specific niche addressed by RAKE ML lacks a dedicated public sizing estimate.

Competitive Landscape

MIXED RAKE ML enters a market where the primary competition is not from direct AI replicas, but from a fragmented ecosystem of incumbent service providers and adjacent technology platforms.

No direct, named competitors building a vertically-integrated causal AI platform for roof failure prediction are identifiable in public sources. The competitive analysis therefore maps a landscape of substitutes and potential future entrants.

• Traditional Inspection & Engineering Firms. Companies like Bureau Veritas or local engineering consultancies provide manual, on-site condition assessments. Their advantage is regulatory acceptance and human expertise, but they are labor-intensive, slow, and lack predictive scale. RAKE ML's value proposition is the automation and forward-looking prediction of these services.
• Geospatial Analytics & Remote Sensing Platforms. Public companies like Planet Labs or Airbus provide high-resolution satellite imagery. Startups like Cape Analytics use computer vision on aerial imagery to extract property attributes for insurance. These firms are data providers or offer descriptive analytics (e.g., "this roof has damage"), but they do not claim to model the underlying physics of degradation over time, which is RAKE ML's stated technical wedge [rakeml.com, retrieved 2024].
• Property Management & Facility Management Software. Platforms like MRI Software or Accruent offer capital planning modules that track repair history. They are workflow systems that rely on manually inputted condition data, creating an opportunity for RAKE ML to act as a data feed, though also a potential threat if they develop or acquire similar predictive capabilities.
• Insurtech Underwriting Platforms. Companies like Zesty.ai (now part of Verisk) use geospatial data to assess property risk for insurers. Their focus is typically on acute perils like wildfire or flood, not the chronic, physics-based degradation of roofing materials. This represents an adjacent competitive segment that could expand into RAKE ML's niche.

RAKE ML's claimed defensible edge rests on its proprietary causal AI models. The company states its platform "understand[s] the fundamental physics of why, when, and how roofs fail" and generates data "by simulating millions of physically-grounded degradation and failure scenarios" [rakeml.com, retrieved 2024]. If validated, this approach could yield more accurate long-term predictions than pattern-matching on imagery alone. This edge is technical and perishable; it depends on continued research lead, the quality and exclusivity of training data from simulations, and the ability to patent core methodologies.

The company's most significant exposure is its lack of commercial footprint and distribution. Established geospatial analytics firms have existing sales channels into large insurers and property managers. A company like Cape Analytics, which has raised over $60 million and counts major insurers as customers, could decide to build or buy a physics-based prediction layer, leveraging its existing distribution to outflank RAKE ML [PitchBook]. Furthermore, RAKE ML's focus on a single component (roofs) of the building envelope may limit its initial addressable market compared to platforms analyzing entire properties.

The most plausible 18-month scenario hinges on proof of commercial adoption. If RAKE ML can secure a flagship partnership with a national insurer or a large real estate investment trust, it would validate its model and create a data feedback loop that accelerates its technical edge. In this case, the "winner" would be RAKE ML, carving out a defensible niche. The "loser" would be the manual inspection consultancies, who would see the highest-margin, predictive part of their service automated first. Conversely, if RAKE ML cannot close enterprise deals and remains in stealth, a well-funded geospatial analytics incumbent is the likely winner, as it would have the time and resources to develop a competing solution organically or through acquisition.

Data Accuracy: YELLOW -- Competitive mapping is inferred from company positioning and analysis of adjacent market segments; no direct competitors are named in public sources.

Opportunity

PUBLIC

RAKE ML's opportunity rests on the chance to become the definitive computational intelligence layer for the world's physical assets, a category that has seen little of the AI transformation that reshaped finance or software.

The headline opportunity is the creation of a category-defining platform for predictive asset health, starting with roofs but extending to the entire built environment. The company's positioning as a "vertically-integrated AI platform" suggests an ambition to own the full stack from data ingestion to actionable workflow, not just sell a point solution [rakeml.com, retrieved 2024]. This outcome is reachable because the core pain point,unplanned, catastrophic asset failure,drives multi-billion dollar annual costs in insurance claims, emergency repairs, and lost asset value. By claiming to understand the "fundamental physics" of failure, RAKE ML aims to move beyond reactive inspection reports to a system of record for asset condition [Perplexity Sonar Pro Brief, retrieved 2024]. The first vertical, roof assessment, serves as a wedge into property insurance and large-scale asset management, two industries with both the budget and the acute need for better predictive tools.

Multiple paths could lead from this wedge to massive scale. The most plausible scenarios hinge on specific commercial catalysts rather than broad market tailwinds alone.

Scenario	What happens	Catalyst	Why it's plausible
Insurance Underwriting Standard	RAKE ML's risk signals become a mandatory input for property insurance underwriting and renewal decisions across major carriers.	A strategic partnership or API integration with a top-10 property & casualty insurer.	The company's public messaging explicitly targets "objective condition evidence to support underwriting and renewals" [rakeml.com, retrieved 2024]. The LinkedIn post about expanding its "commercial pipeline with Risk Signal Solutions" indicates active commercial outreach in this direction [LinkedIn, May 2025].
Asset Management Operating System	The platform becomes the central planning tool for large real estate portfolios (REITs, institutional owners), dictating capital expenditure timing and reserve studies.	A flagship deployment with a publicly-traded REIT or a major pension fund's real estate arm.	RAKE ML's product claims directly address "capital planning and asset management" by translating condition signals into repair timing [rakeml.com, retrieved 2024]. The built environment lacks a dominant software player for predictive maintenance, leaving a gap for a vertically-integrated solution.

A successful wedge into either scenario would create a compounding advantage. In the insurance case, each new policy assessed would generate proprietary loss data correlated with RAKE ML's initial predictions, refining the causal models and creating a data moat that pure inspection firms cannot match. For asset managers, the platform's value increases as it ingests data across a portfolio, enabling comparative analytics and early warning systems that become harder to displace. The company's claim to generate data "by simulating millions of physically-grounded degradation and failure scenarios" suggests a foundation for this flywheel, where simulated and real-world data continuously improve model accuracy [rakeml.com, retrieved 2024].

The size of the win, should one of these scenarios play out, can be framed by looking at comparable companies that digitized analog, inspection-heavy processes. For instance, Katerra, despite its operational challenges, reached a private valuation of over $4 billion on the premise of vertically integrating construction. A more direct, though smaller, precedent might be EagleView, a provider of aerial imagery and analytics for the roofing and insurance industries, which was acquired for roughly $650 million in 2015. If RAKE ML's causal AI platform could capture even a fraction of the global property insurance risk assessment market,or become the standard tool for a segment of commercial real estate asset management,a valuation in the hundreds of millions to low billions is a plausible outcome (scenario, not a forecast). The total addressable market is anchored by the sheer volume of physical assets; one industry report estimates the global property insurance market alone exceeds $1.6 trillion in direct premiums written [Swiss Re Institute, 2023], representing a massive pool of risk assessment spend.

Data Accuracy: YELLOW -- The opportunity analysis is built on the company's stated product claims and target markets, which are publicly documented. The scenario plausibility is supported by these claims and evidence of commercial activity. Market size comparables are drawn from independent industry reports, but the linkage to RAKE ML's specific potential is an analyst inference.

Sources

PUBLIC

1. [rakeml.com, retrieved 2024] RAKE ML , Predict & Prevent Asset Failure Before It Happens | https://www.rakeml.com

2. [LinkedIn, May 2025] RAKE ML's Post - startups #founders #ai #builtenvironment | https://www.linkedin.com/posts/rake-ml_startups-founders-ai-activity-7413987709838311426-CXjB

3. [LinkedIn, retrieved 2024] RAKE ML | https://www.linkedin.com/company/rake-ml

4. [PitchBook, retrieved 2024] RAKE ML 2026 Company Profile: Valuation, Funding & Investors | PitchBook | https://pitchbook.com/profiles/company/756737-47

5. [Perplexity Sonar Pro Brief, retrieved 2024] RAKE ML Company Brief | Not available

6. [Swiss Re Institute, 2024] sigma 1/2024 - World insurance: stirring up the future | Not available

7. [MarketsandMarkets, 2023] Commercial Real Estate Software Market by Component, Deployment, Application, End User and Region - Global Forecast to 2028 | Not available

8. [National Oceanic and Atmospheric Administration, 2024] U.S. Billion-Dollar Weather and Climate Disasters | Not available