> ## Content Index > Fetch the complete content index at: https://intelligence.artificialrealitymedia.com/llms.txt > Use this file to discover other available public pages before exploring further. # Four permitting startups, and the half none of them solve - URL: https://intelligence.artificialrealitymedia.com/permitting-software-solves-the-wrong-half/ - Published: 2026-09-02T12:09:00.000Z - Updated: 2026-09-02T12:09:00.000Z - Description: PermitFlow, GreenLite, Pulley and Symbium automate filing. One restaurant operator lost $30,000 to the step before filing, and nobody sells it. - Author: Galina Fendikevich - Tags: Restaurants, #For AI Companies, #What needs to be built, #Import 2026-09-03 05:09 Permitting software has automated the filing. The failure that cost one restaurant operator $30,000 happened before filing. [PermitFlow](https://permitflow.com/?ref=intelligence.artificialrealitymedia.com), GreenLite, [Pulley](https://withpulley.com/?ref=intelligence.artificialrealitymedia.com) and [Symbium](https://symbium.com/?ref=intelligence.artificialrealitymedia.com) prepare, submit, manage or encode permit requirements, and none of them tells a first-time restaurateur that a mechanical, electrical and plumbing engineer has to come first. Dr. Nikki Siso, founder of Conscious Kitchen outside Austin, lost a month finding that out. ## What did the operator actually fail at? Not filing. She filed. Siso went to the planning permit department first, waited two weeks for a meeting, waited another two weeks for an engineer, and paid $30,000 to a contractor she concluded afterwards she had not needed: *"I was one person away from doing it all by myself."* The knowledge she was missing fits in five words: start with the MEP engineer. She asked AI for the sequence and got the health department's application link back. In her account, *"for some reason it didn't know: start with the MEP."* ## What do the funded products in this category do? They solve permit preparation and submission across jurisdictions — a genuinely hard problem, and not the one she hit. Artificial Reality identified four as apparently relevant to the requirements below. None was named or endorsed by the operator. - **PermitFlow** — described as TurboTax for permitting: AI-assisted permit preparation, submission and tracking across thousands of municipal jurisdictions, restaurant build-outs included. Series B, $54M, December 2025. - **GreenLite** — AI permitting. Series B, $49.5M, September 2025\. Artificial Reality's research on this company is thin beyond stage and funding. - **Pulley** — permitting software plus an expert service that researches jurisdiction requirements and manages submittals for commercial projects, including restaurant tenant improvements. Seed, $4.4M. - **Symbium** — computational compliance, encoding local regulations so requirements can be determined programmatically. Early stage, BBG Ventures. Artificial Reality's detail on scope here is limited. The category is construction-permit-focused. Artificial Reality found no product covering health department inspection preparation. ## What would the missing product have to do? Six requirements come from the operator. Three are additions from Artificial Reality, the AI buyer-evidence podcast hosted by Galina Fendikevich. Stated by Siso: give the correct order of operations, starting with which professional to engage first; be genuinely local at municipality level rather than city or state; include what inspectors actually check on site, not only what departments publish; cover the required physical documentation, including signage that must be posted; help complete the applications themselves, which was Fendikevich's suggestion in the interview and which Siso endorsed — "100%"; and produce a pre-inspection spec sheet, which in her words "would have been gold." Artificial Reality's additions: cover health and food-service permitting rather than construction alone; flag when a step can be self-performed, since she hired a $30,000 contractor partly because nothing told her she did not need one; and warn about jurisdiction boundaries, because an address near a city line can fall under a dramatically lighter regime. ## Why has nobody built the sequence layer? Because the knowledge has no source to ingest. It is not on a government website. It sits in the working memory of people who do build-outs, and there is no dataset to license. Encoding it means either interviewing practitioners jurisdiction by jurisdiction or capturing it from projects as they happen. Both are slow, unglamorous and expensive relative to the addressable market of independent restaurant openings. The economics are perverse in a specific way. **The information is worth a great deal to each person who needs it, and each person needs it exactly once.** No individual will pay to compile it, and nobody has yet found a way to make compiling it pay. ## What is the tractable version of this? The pre-inspection spec sheet, and it is the more valuable of the two gaps. Health departments inspect the same things repeatedly. Any restaurateur who has been through an inspection knows the list — refrigerator temperatures, water reaching at least 110 degrees, the signage that has to be on the wall. Nothing about aggregating that is technically hard. It has simply never been anyone's job. The cost of its absence, in this case, was a $115 water heater discovered on the morning it failed her, after a plumber had quoted $250 for the same fix. The second tractable feature is the jurisdiction warning. Siso found Austin's requirement list roughly ten times longer than neighbouring Westlake Hills'. Any product that already maps municipalities knows this and does not surface it. ## What should a builder in this category do next? Run one test on your own product. Ask it what a first-time restaurant operator should do first in a named municipality. If it answers with an application, you are solving the second problem well and the first one not at all — and the first one is the part worth $30,000 to the person asking. ## Where this comes from S1E6: I trusted AI to build my restaurant, it cost me $30,000 — the full interview with Dr. Nikki Siso. Listen or watch: [YouTube](https://www.youtube.com/watch?v=7LF4Ro8Hg6M&ref=intelligence.artificialrealitymedia.com), [Spotify](https://open.spotify.com/episode/5TwX8rLnDKkne9REQjFCEi?ref=intelligence.artificialrealitymedia.com) or [Apple Podcasts](https://podcasts.apple.com/us/podcast/i-trusted-ai-to-build-my-restaurant-it-cost-me-%2430-000/id6780699025?i=1000777268211&ref=intelligence.artificialrealitymedia.com).