MirageDoodle, Inc.
AI-native workspace for municipal development review
About
At MirageDoodle, we’re building coordinated AI agents that live inside cities’ cross-department development review workflows—doing the heavy lifting, escalating judgment calls, and letting staff focus where human expertise actually matters. MirageDoodle provides expert-level guidance in minutes based on precedents, helping cities and design firms navigate an increasingly complex, fragmented regulatory environment We serve as strategic partners to forward-thinking cities and design firms across the country.
Founders
Founder
Lin combines a background in Urban Planning and HCI (Drop-Out) from UMich, bringing policy insight and design expertise to MirageDoodle. He leads product strategy with a focus on zoning and compliance, with extensive experience in code analysis, density studies, and scenario modeling. An APA Planning and Design Competition Honorable Mention recipient, Lin bridges technical and policy worlds to make compliance faster, clearer, and more accountable.
Founder
A Georgia Tech graduate, he has worked on autonomous cars, boats, and robotic systems, consistently pushing the boundaries of autonomy. George is known for rapidly engineering novel, highly cost-effective solutions—bridging advanced AI with real-world applications from self-driving vehicles to scalable automation.
AI Research Report
Problem & Solution
Problem/Solution Report
The core problem MirageDoodle targets is the growing complexity and fragmentation in zoning, building codes, and cross‑departmental development review processes at U.S. cities and counties. Traditional plan review spans multiple departments (planning, building, fire, engineering) and relies on manual QA/QC and code lookups, creating bottlenecks, long approval times, and high costs. MirageDoodle’s materials frame this as both a regulatory and operational challenge that delays development and burdens staff.
MirageDoodle’s proposed solution is an AI‑powered platform that embeds “coordinated AI agents” directly into cross‑departmental development review workflows. The product highlights a few foundation components: Plan & Document Intelligence (turning complex AEC drawings, markups, and schedules into structured knowledge), a Code & Precedent Graph (capturing zoning, fire, and engineering codes with real‑world precedent), a Multi‑Agent Review Engine (triaging, checking, and escalating items that require human judgment), and Human‑in‑the‑Loop collaboration (smart handoffs across teams). The platform aims to provide instant analysis that is domain‑tuned to specialized diagrams and local codes, and to execute agentic compliance checks across zoning and building regulations.
The company emphasizes measurable value: faster reviews, higher automation, and meaningful cost reduction. Specifically, site messaging claims up to 90% faster reviews, 95% automation, 24/7 agentic processing, and up to 80% cost reduction — while keeping final judgment and accountability with human reviewers. YC’s profile describes the positioning succinctly: an AI‑native workspace for municipal development review that escalates judgment calls and lets staff focus where human expertise matters. The blog narrative further asserts that as construction accelerates, approvals must keep pace, reinforcing the urgency of automation in development review.
Market & Competitors
Market & Competitors Report
MirageDoodle operates within the govtech segment of permitting, licensing, and compliance — specifically the sub‑domain of municipal development review, plan review, and code/zoning compliance. A broad set of vendors offers enterprise permitting & licensing platforms that digitize workflows and citizen front doors. Accela, OpenGov, and Tyler Technologies (EnerGov) are among the most established providers, with strong installed bases across city and county governments. Their propositions typically stress configurable forms/workflows, online portals, inspection mobility, GIS integration, and back‑office automation.
Accela positions the Civic Platform and “out‑of‑the‑box civic applications” as highly configurable permitting, licensing, and code enforcement solutions with end‑to‑end support. OpenGov markets a modern, configurable permitting & licensing system with a “virtual front desk,” GIS integration, mobile inspections, and plan review integrations to speed approvals. Tyler’s EnerGov brochure highlights a GIS‑centric, mobile‑first suite for managing permitting, licensing, and regulatory operations in one centralized system. These providers digitize processes and data flows, but are not primarily positioned as AI‑agent platforms for the substantive technical review of plans and codes.
On the code and zoning intelligence side, UpCodes offers “Copilot,” an AI‑powered assistant to answer code questions, perform calculations, and provide assembly recommendations based on selected jurisdictions and codebooks. Gridics provides cloud‑based zoning applications for municipalities and real estate professionals, focusing on parcel‑specific zoning and property analysis. These adjacent offerings speak to growing demand for AI‑assisted code interpretation and zoning insights, but they are generally not marketed as multi‑agent systems running end‑to‑end plan reviews inside municipal workflows.
MirageDoodle differentiates via an “AI‑native workspace” that couples plan/document intelligence with a code & precedent graph and a coordinated multi‑agent review engine built for cross‑department development review. If the platform’s claimed speed and automation translate into production environments, it could complement (or in some cases compete with) enterprise permitting suites by handling the technical review depth more autonomously, while still integrating with human reviewers and existing systems. Competitive risks include incumbent platform lock‑in, procurement cycles, data availability/quality, and the need for rigorous accuracy and auditability in public sector decisioning.
Total Addressable Market
Quantitative TAM Report
Top‑down market sizing suggests the global “permitting and licensing software” category ranges from roughly sub‑$1B to mid‑single‑digit billions annually, depending on definition and scope. For government‑specific permitting and licensing software, one market analysis pegs the global market near ~$1.0B in 2025 with a CAGR of ~6% through 2033. Broader market trackers with looser category definitions cite materially larger figures — for example, estimates around $5.8B globally in 2024 with double‑digit CAGR to 2033, and some reports that appear to include wider enterprise licensing domains suggest ~$65B in 2023. These disparate figures underscore differences in segmentation (government vs. broad enterprise), included applications (permitting, licensing, compliance, inspections, environmental, utilities, etc.), and methodology.
A bottom‑up reference point for the U.S. is the number of relevant governmental units. The 2022 Census of Governments counted 90,837 governments (including 3,031 county governments and 35,705 municipal/township governments). Not all issue building permits or conduct plan reviews directly, but the magnitude shows a large base of potential public‑sector customers. On the demand side, U.S. new residential building permits alone average roughly 1.1‑1.5 million units SAAR (e.g., 1.312M SAAR in Aug 2025), and large jurisdictions also process substantial commercial permits. These volumes stress existing processes and motivate adoption of digital and AI‑enabled review.
Framing MirageDoodle’s initial wedge as “municipal development review and plan/code compliance” (a subset of permitting & licensing software), a conservative working TAM band can be derived by segmenting broader market estimates to technical plan review use cases and relevant buyers (municipalities, counties, and select design firms). Using the narrower government‑only market (~$1.0B) and attributing ~25‑35% to development review/plan review and code/zoning compliance yields an initial global TAM of approximately $250‑$350 million. Alternatively, applying a similar 25‑35% share to the broader $5.8B estimate implies a wider band of ~$1.5‑$2.0 billion, reflecting inclusion of more adjacent compliance workflows and geographies. A U.S.‑only, bottom‑up sanity check — assuming ~15,000‑20,000 permit‑issuing jurisdictions and an average annual spend of ~$40‑$80k on plan review/compliance software and AI — yields ~$0.6‑$1.6 billion, broadly consistent with the bands above.
Methodology: triangulate published global market estimates for permitting/licensing software (noting scope differences), anchor addressable buyer counts using official government unit statistics, and apply conservative allocation shares to isolate the development review/code compliance slice most aligned with MirageDoodle’s product. Given variability in published figures and evolving AI capabilities in government, these TAM bands should be revisited as deployment models, pricing, and adoption rates clarify.
Founder Analysis
Founders Background Report
MirageDoodle, Inc. was founded by Xuanshu Lin and George Zhai. The company publicly describes itself as building an AI‑native workspace for municipal development review, participating in the Y Combinator Winter 2026 batch. Both founders are profiled on the company’s website and on Y Combinator’s company page, which highlight complementary strengths across urban planning, HCI, AI, and robotics.
Xuanshu Lin is a co‑founder who brings a background in Urban Planning and Human‑Computer Interaction (HCI) from the University of Michigan. The company’s About page and YC profile note that Lin focuses on zoning and compliance with experience in code analysis, density studies, and scenario modeling, and that he received an Honorable Mention in the APA Planning and Design Competition. Lin’s public LinkedIn content corroborates honors and awards, showing planning‑related recognition and academic achievements. These experiences position Lin to translate planning policy and code interpretation into product strategy for AI‑assisted compliance workflows.
George Zhai, co‑founder and the company’s designated copyright agent, is a Georgia Tech graduate known for applied AI and robotics work across autonomous cars, boats, and robotic systems. The company’s About page and YC profile emphasize Zhai’s track record of rapidly engineering cost‑effective solutions and bridging advanced AI to real‑world applications. MirageDoodle’s Terms of Use lists his contact details for the company, reinforcing his formal corporate role and responsibility.
Together, the founders’ profiles underscore a blend of domain expertise (urban planning, zoning, and compliance) and technical depth (AI, agent systems, and autonomy). Their YC participation and early accelerator backing further indicate a focus on building and productizing a specialized AI platform for development review inside government workflows.
Unlock Full AI Research Report
Enter your email to access the complete analysis.
We'll never spam you. Unsubscribe anytime.