Many U.S. businesses—especially those importing hardware, electronics, or coated textiles—are unaware of a major EPA reporting deadline coming up on April 13, 2026.
Under TSCA Section 8(a)(7), the EPA is requiring a one-time report on any PFAS (per- and polyfluoroalkyl substances) manufactured or imported into the U.S. since January 1, 2011. This includes "article importers," meaning if you imported a product containing a PFAS-coated component at any point in the last 15 years, you may be required to report it.
I built pfasreporting2026.com to act as a triage tool for small-to-medium businesses that don't have a dedicated environmental compliance team. It features a 2-minute assessment to help you determine if you're in scope and provides a "Supplier Demand Letter Kit" (in English and Mandarin) to help you get the necessary data from your international suppliers.
The penalties for non-compliance are significant (up to $37,500 per day), and the lookback to 2011 makes data collection a serious hurdle. I’d love to hear your thoughts on the tool or any questions about the regulation itself.
Fight for Justice with Mathematical Precision
AI-powered sentencing audits, legal document generation, and case analysis for Federal, Military, ICE, and State systems
Fight for Justice with Mathematical Precision
AI-powered sentencing audits, legal document generation, and case analysis for Federal, Military, ICE, and State systems
Or should I stick to the crowdfunding option on the platform to purchase the land? All contributors would get some sort of lifelong benefits or profit share.
Hey HN,
I recently launched YachtVest (hosted at yachtvaluereport.com) to bring financial rigor to an industry that is notoriously emotional and opaque: superyacht acquisitions.
Buying a vessel is often treated like buying a very expensive car, but financially, it behaves more like a complex depreciating asset with massive opex. I realized that while there are plenty of "Zillow for boats" listings sites, there wasn't a "Bloomberg Terminal" for analyzing the asset itself.
What it does:
Depreciation Modeling: We use historical data from ~15,000 sales to project value curves 5 years out.
True Cost of Ownership (TCO): A calculator that includes the stuff brokers often gloss over—slip fees, crew salaries, insurance, and fuel consumption curves.
Risk Detection: We use AI to scan owner forums and service bulletins to flag model-specific issues (e.g., "known hull delamination issues on 2019 models").
Hello HN,
I built Compare Contractor Quotes to solve a problem I faced during my own home renovation: trying to compare three different quotes that looked nothing alike. One was a detailed PDF, one was a Word doc, and one was essentially a photo of a napkin.
It was nearly impossible to see if they were quoting the same scope of work. I realized that the "cheapest" quote was actually just leaving out expensive items (like debris removal and permits) that would have hit me as change orders later.
What it does:
The tool takes unstructured contractor bids (PDFs, images, scans) and uses LLMs to parse them into a structured data format. It then:
Normalizes the data: Maps different terminologies to a standard scope of work.
Identifies Gaps: Highlights if one contractor included a line item that others missed.
Flags Risks: Checks for front-loaded payment schedules or vague language that puts you at risk.
How it works:
We use a combination of OCR and a fine-tuned LLM pipeline to extract line items, prices, and terms from the raw documents. The challenging part was getting the model to understand the context of construction lingo (e.g., differentiating between "rough-in" and "finish" plumbing) across different trades.
The Goal:
To give homeowners the leverage of a General Contractor. By standardizing the bids, you can see exactly where the price differences are coming from and negotiate from a position of data rather than guessing.
It’s live now at https://comparecontractorquotes.com.
There is a free preview mode so you can test it out without paying. I’d love to hear your feedback on the extraction accuracy and if there are specific "red flags" you think I should add to the detection logic!
Hey HN,
I built nycnoisecameraticket.com to help drivers fight the new automated noise tickets in NYC.
The Engineering Challenge:
NYC is deploying microphone arrays to pinpoint loud cars. But NYC is also a "concrete canyon." Sound reflects, reverberates, and amplifies.
The Dispute Strategy:
My tool generates defenses based on the limitations of acoustic triangulation in complex environments.
The "Canyon Effect": Was the reading distorted by a nearby glass building or construction wall?
Ambient Interference: Was there a siren, jackhammer, or subway grate rumble nearby?
Source Ambiguity: How does the camera distinguish your engine from the modified Harley in the next lane?
If the city wants to fine people $800 via algorithm, the algorithm needs to account for physics. My tool helps users demand that proof.
Hey HN,
I built workzonespeedingticket.com because automated cameras are issuing fines in construction zones even when no work is happening and no workers are present.
The Issue: Many states (like NY and PA) are rolling out 24/7 monitoring. The cameras are dumb; they don't know if the construction crew went home for the weekend or if the "Work Zone" signs were left up by mistake. If you drive the normal limit, you get a $50-$100 ticket in the mail.
The Fix: I built a workflow that helps you draft a "Statement of Facts" disputing the "Active" status of the zone.
How it works:
You input the ticket time/location.
The tool generates a dispute letter challenging the state to prove the zone was active (required by law in many jurisdictions).
It formats it for the specific county clerk or administrative judge.
It’s a "David vs. Goliath" tool. If the state is going to automate the fine, we should automate the defense.
Hey HN,
I launched nycnoisecameraticket.com to help New Yorkers automate disputes for the city's new "SoundVue" noise cameras.
The Context: NYC has started deploying microphone arrays (acoustic cameras) that triangulate sound sources. If a vehicle emits >85dB at 50ft (roughly the volume of a lawnmower), it triggers a camera that snaps your plate. The problem? The fines are massive ($800–$2,500), and even stock vehicles (like standard Porsches or Jaguars) are triggering them just by accelerating normally.
The Problem: The "Notice of Liability" is often vague, but the defenses are technical (calibration logs, microphone distance certification, ambient noise interference). Most people just pay the $800 because hiring a traffic lawyer costs more.
The Solution: I built a specialized LLM workflow that:
Ingests the ticket details (location, time, alleged dB reading).
Generates a specific legal defense letter citing the NYC Noise Code (Section 24-237) and requesting the specific calibration records for that camera unit.
(In progress) Automates the physical mailing via the Lob API so you never have to touch a mailbox.
The stack is React/Lovable on the front end + Supabase. I'm looking for feedback on the "stock vehicle" defense logic if anyone here has experience with acoustic engineering or NYC administrative law.
Hey HN,
I built WorkZoneSpeedingTicket.com to solve a specific problem: automated work zone speed cameras are popping up everywhere (NY, CA, PA, etc.), and the fines (~$50-$100) are in that awkward "too cheap to hire a lawyer, but too expensive to just roll over and pay" range.
The Problem: States are rolling out automated enforcement in construction zones. Often, the signage is non-compliant, or the "work zone" is inactive, but the cameras issue tickets anyway. Disputing them manually is a hassle designed to make you pay.
The Solution: I built a tool that takes your ticket details and generates a jurisdiction-specific dispute letter citing relevant vehicle codes (e.g., California Vehicle Code or NY VTL).
How it works:
User inputs ticket info.
AI cross-references the location/time against state-specific defense templates (signage distance, equipment calibration requirements).
Generates a PDF dispute letter.
(Coming soon) We automate the certified mailing via the Lob API so you don't have to go to the post office.
The goal is to democratize access to legal disputes for low-level infractions. I used Lovable to spin up the frontend quickly and am refining the backend logic now.
I’d love feedback on the flow or if anyone knows specific state statutes I should add to the database next.
Hi HN,
AeroGrid is an attempt to standardize the "wild west" of low-altitude airspace before it gets too crowded.
We've mapped out a 3-layer grid (Delivery, Passenger, Cargo) across 50 states and are releasing a set of open standards (AUAS-100 to 900) for cities to adopt. The idea is to create "digital highways" in the sky with defined on-ramps, speed limits, and right-of-way rules.
The site includes a flight simulator and a compliance framework that (controversially) proposes protocols for "kinetic interdiction" of non-compliant drones.
Would love feedback on the grid topology or the proposed standards.
Hey HN, OP here.
I built FightHOAFines.com because I got tired of the arbitrary enforcement and bureaucratic nightmare that is the modern Homeowners Association.
The Problem:
HOA boards often send violation notices based on vague interpretations of bylaws or without following proper state notification procedures. Most homeowners just pay the fine because researching the CC&Rs (Covenants, Conditions, and Restrictions) and drafting a formal response takes too much time.
What this tool does:
You upload your violation notice (or describe the issue), and the tool generates a formal dispute letter. It doesn't just complain; it structures a "legal-lite" argument.
Under the Hood:
Input: It takes the violation details and (optionally) your specific HOA documents.
Processing: I'm using [Insert Model, e.g., GPT-4o/Claude] to parse the legalese of the violation.
Retrieval: The system cross-references the violation against a database of common defenses and specific state statutes (currently optimized for [Your State/Florida] but working on others) to find procedural errors the HOA might have made.
Output: It generates a formatted PDF letter ready to mail.
Why I built it:
I realized that 90% of fighting an HOA is just sounding professional enough and citing the right statute to make them back down. I wanted to democratize that "lawyerly pushback" for people who can't afford actual counsel for a $100 trash can fine.
I’d love feedback on:
The letter quality, is it too aggressive or too passive?
Any specific edge cases (e.g., selective enforcement) you think I should add.
Thanks!
I built RepoCard because I kept running into a friction point every time I shipped a side project: I’d spend hours coding, but when it came time to share it on Twitter or here, I’d lose 45 minutes fussing around in Figma trying to make a decent "social preview" image.
I realized most of the data I needed for the marketing assets was already in the repo (description, language, stars, code snippets).
RepoCard (https://repocard.dev) is a "launch kit" that auto-generates these assets from a single GitHub URL.
What it generates:
1. Social Cards (OG Images): Styled previews with your repo stats.
2. CodeSnaps: Aesthetic screenshots of your actual code (fetches from main/index files).
3. Launch Posts: AI reads your README and drafts a release thread.
4. File Trees: ASCII art of your project structure for documentation.
It’s free to try (no signup required). There is a $9 one-time lifetime deal that unlocks high-res exports, removes watermarks, and adds premium themes (Cyberpunk, Swiss, HN Orange).
--
Anticipating a few questions:
1. How does it work?
It uses the GitHub Public API to fetch repo metadata and raw file content. The "Launch Post" feature uses a lightweight LLM to summarize the README. Everything is built on Next.js, Supabase, and Stripe.
2. Why not open source?
I’m a solo dev and this is a paid product to support my time, so the source is closed for now. However, I’m using standard libraries (html-to-image, shiki for syntax highlighting) if you want to build your own version.
3. Do you store my code?
No. We fetch the file content to render it on the canvas in your browser. We do not store or train on your repository code.
4. Why $9?
I’m trying to avoid the subscription fatigue we all hate. A one-time price covers the API/LLM costs and feels fair for a utility tool you might use 5-10 times a year.
I’d love to know which templates you prefer or if there are other assets (maybe a GIF generator?) that would be useful for your launches.
OP here. I built this because the standard property tax appeal process feels broken—you either spend hours deciphering government forms yourself or hire a firm that takes 50% of your first year's savings.
Under the hood, the app ingests property data (assessments, recent sales) and uses LLMs to draft arguments based on local statutes. For the fulfillment layer, I integrated the Lob API to handle the printing and USPS Certified Mail so users don't have to visit the post office.
I’m currently testing a flat-fee model ($29/$49) rather than the industry-standard contingency fee. I'd love to hear your feedback on the pricing structure or the "comparable sales" search flow.