Pound Puppies to guide dogs

Cj Jones — Mon, 17 Aug 2026 01:57:03 +0000

This is a submission for Weekend Challenge: Dog Days Edition

Pound Puppies to Guide Dogs for the Blind: Closing the $50K Gap with Open ADI Standards

What I Built

I built Pound Puppies to Guide Dogs for the Blind — a single-file, offline-capable web application that turns any competent dog trainer into an ADI-standard guide dog trainer by giving them the curriculum, tools, and legal framework that accredited programs use.

The problem is well-documented:

$50,000–$70,000 is what it costs accredited programs to produce one guide dog team [^1]
2–4 years is the typical wait time for a program-placed guide dog [^2]
1.3 million visually impaired Americans are waiting
50% of dogs bred specifically for guide work still wash out

Meanwhile, shelter puppies with ideal temperaments have no pathway into guide work. The curriculum exists. The ADA standards are public. The trainers exist. But the bridge between them was missing.

This app is that bridge.

It includes:

A 6-module ADI-based training curriculum covering SAITE screening through the Public Access Test
Voice-guided training commands (hands-free for trainers working with dogs)
An interactive ADA Compliance Checklist with persistent local storage
A Progress Tracker for logging milestones and handing off to evaluators
A Digital Liability Waiver protecting volunteer trainers and rescue organizations
Full screen-reader optimization, high-contrast mode, and offline functionality

Demo

🔗 Live site: https://chrisj1s712-cmyk.github.io/Pound-Puppies-to-Guide-Dogs-for-the-Blind/

The app is a single HTML file. No build step. No server. Works entirely in the browser with localStorage for persistence. Open it, add it to your home screen, and it behaves like a native app — even without a connection.

Code

📁 GitHub Repo: github.com/chrisj1s712-cmyk/Pound-Puppies-to-Guide-Dogs-for-the-Blind

The entire application is one self-contained index.html. This was a deliberate architectural choice: the target users (volunteer trainers, foster networks, and visually impaired handlers) need something that deploys anywhere without dependencies.

How I Built It

The Builder's Path

I am a braille transcriber by trade. That work taught me something critical: accessibility is not a feature you bolt on at the end. It is the foundation everything else rests on. When you spend your days ensuring that printed information survives the transition to tactile form, you develop a intolerance for interfaces that exclude.

That path led me to build aurastoria.com — an accessibility center focused on getting adaptive technology and inclusive design resources into the hands of people who need them. It also led to pifcoin.org, which I originally built to get smart glasses with AI technology into the hands of my local chapter of the National Federation of the Blind. The goal was never charity. It was tool parity: making sure that the same technological leverage available to the sighted world reaches the blind and visually impaired community without a $50K price tag or a three-year wait.

This challenge is perfect for what I do because every build I ship — whether it's a hardware funding platform, an accessibility resource hub, or a guide dog training curriculum — is governed by the same imperative: the impairment of the user must never be the barrier to the tool.

The Gap Analysis

I started by researching why the guide dog bottleneck persists. ADI-accredited programs do extraordinary work, but they're capacity-constrained. The Guide Dog Foundation states it costs more than $50,000 to produce a single team [^1]. Waitlists stretch 2–4 years [^2]. And despite all that investment, breeding programs still see roughly half their candidates wash out.

The ADI Standards themselves are rigorous and peer-reviewed — they're the benchmark for excellence in the industry ¹. But those standards have historically been locked behind organizational walls. My goal was to open-source the pathway, not the dog.

Curriculum Architecture

I structured the training into six modules mapped to ADI's documented phases:

Selection & Assessment — SAITE protocol, vet screening, socialization baseline
Basic Obedience — Sit, stay, recall, loose-leash walking per IAADP's 120-hour minimum standard ²
Public Access Manners — Elevators, food refusal, under-chair settling, bathroom on command
Advanced Guide Skills — Intelligent disobedience, curb stops, obstacle navigation, platform boarding
Handler Matching — Needs assessment, lifestyle compatibility, 2-week bonding period
Certification — ADI Public Access Test documentation

Each module is collapsible, screen-reader annotated, and cross-referenced with the ADA's two-question rule for public access.

Voice & Accessibility

For voice-guided commands, I evaluated ElevenLabs. I have an active subscription, but ElevenLabs' cardinal rule for frontend builds is do not expose API keys in client-side code. Rather than violate that policy or build a backend I didn't need, I integrated Gemini for the voice pipeline — honoring the platform's security model while still delivering hands-free audio guidance.

Accessibility wasn't a feature. It was the foundation:

High-contrast mode always available
Semantic HTML for screen readers
No horizontal scroll
Mobile-first responsive layout
localStorage persistence so data survives offline

Legal Framework

The ADA guarantees public access rights without requiring certification or documentation ². But trainers and rescues still need liability protection. I built a digital waiver that acknowledges the inherent risk of working animals (the 50% washout rate is documented even in professional programs) while preserving the recipient's right to choose independence over dependence.

Voices of the Community

"I waited two years for my guide dog. After four months, she washed out. I had to start the wait all over again. Another year gone."
— Sarah M., blind handler

"The school told me 18 months. It's been three years. I just want to walk to the store without asking my neighbor."
— James T., visually impaired, on waitlist since 2023

"My rescue dog has the perfect temperament, but no one would train him for guide work. I had to figure it out alone."
— Maria L., self-trained handler

"I can't afford $70,000. I can't wait three years. But I can adopt a puppy and I have time to train."
— David R., prospective handler

Prize Categories

I am submitting this project for Best Use of Google AI.

I evaluated ElevenLabs for the voice command pipeline — I maintain an active subscription and their TTS quality is industry-leading. However, ElevenLabs' explicit policy prohibits exposing API keys in frontend builds. Rather than compromise on security or spin up a backend server for a single endpoint, I chose Gemini to power the voice-guided training commands. This kept the app a true single-file, zero-backend deployment while respecting platform security policies.

The Bottom Line

Professional guide dogs are life-changing. The organizations that produce them are doing heroic work. But demand outstrips supply by an order of magnitude.

This app doesn't replace accredited programs. It expands the pipeline by giving shelter dogs, volunteer trainers, and waiting handlers the same standards, curriculum, and legal framework that $50K programs use — at the cost of a shelter adoption fee and a trainer's time.

The standards are public. The need is documented. The only missing piece was the delivery mechanism.

Now it exists.

[1]: Guide Dog Foundation — "It costs more than $50,000 to produce one guide dog team" (2024)
[2]: Guide Dogs of America — "Most waitlists average 2-4 years"
My use of AI is for framework,coding, auditing, and organizing. Although I went to school for computer science I found it much easier to use AI for coding it's more efficient. However, to use AI I understand it's limitations and understand tables are necessary, as well as, summarizations and a clean workspace is a necessity prevent fragmenting to protect its precious memory. AI when used properly provides a blueprint not just answers. And I as a developer choose to use.

Code

chrisj1s712-cmyk / Pound-Puppies-to-Guide-Dogs-for-the-Blind

Pound Puppies to Guide Dogs for the Blind trains shelter puppies to ADI standards with full ADA compliance — at a fraction of the cost and a fraction of the wait. Built mobile-first for any dog trainer and any blind handler who refuses to spend 3 years and $70,000 waiting.

Pound-Puppies-to-Guide-Dogs-for-the-Blind

Pound Puppies to Guide Dogs for the Blind trains shelter puppies to ADI standards with full ADA compliance — at a fraction of the cost and a fraction of the wait. Built mobile-first for any dog trainer and any blind handler who refuses to spend 3 years and $70,000 waiting.

View on GitHub

Assistance Dogs International — "ADI Standards have become the benchmark for measuring excellence in the assistance dog industry" ↩
IAADP — Minimum Training Standards for Public Access, 120+ hours requirement; ADA Title II/III service animal provision. ↩

Developers you build for the future...Don't forget to be inclusive!

Cj Jones — Sun, 16 Aug 2026 01:42:36 +0000

Developers you build for the future...Don't forget to be inclusive!

I build things for blind people. I'm a braille transcriber. I work in accessibility every day.

Last January, at an NFB chapter meeting, a member walked in wearing Meta Ray-Ban smart glasses. AI. Camera. Voice. Every blind person in that room wanted to know: How do we get these?

So I built pifcoin.org — a charity crypto to fund assistive tech access. I built aurastoria.com to generate revenue and give the token real liquidity. I built an Accessibility Center into the site because, shockingly, accessibility should be accessible.

Then marketing agencies evaluated my work.

They talked SEO. Funnels. Viral loops. Social traction.

They never asked about the blind users. They never asked about the NFB chapter. And they completely missed the Accessibility Center.

On a website built by an accessibility professional. For an accessibility mission.

Here's what I need from you, developer to developer:

Stop treating accessibility like a feature ticket you close in Sprint 3. It's not a checkbox. It's architecture. If you're building AI tools, camera apps, voice interfaces, or literally anything with a screen — blind users are not an edge case. They're your users.

The Meta Ray-Bans exist right now. The software exists right now. The only gap is the pipeline between "built" and "in their hands."

If you're a developer who wants to build accessible AI tools, or you know how to bridge that gap, I want to talk. Not "let's circle back." Not "I'll put it on the roadmap." Now.

Email me at chrisj1s712@gmail.com. Or drop a comment below. But don't just upvote and scroll. That's how we got here.

DEV Community: Cj Jones

Pound Puppies to guide dogs

Pound Puppies to Guide Dogs for the Blind: Closing the $50K Gap with Open ADI Standards

What I Built

Demo

Code

How I Built It

The Builder's Path

The Gap Analysis

Curriculum Architecture

Voice & Accessibility

Legal Framework

Voices of the Community

Prize Categories

The Bottom Line

Code

chrisj1s712-cmyk / Pound-Puppies-to-Guide-Dogs-for-the-Blind

Pound Puppies to Guide Dogs for the Blind trains shelter puppies to ADI standards with full ADA compliance — at a fraction of the cost and a fraction of the wait. Built mobile-first for any dog trainer and any blind handler who refuses to spend 3 years and $70,000 waiting.​

Pound-Puppies-to-Guide-Dogs-for-the-Blind

Developers you build for the future...Don't forget to be inclusive!

Pound Puppies to Guide Dogs for the Blind trains shelter puppies to ADI standards with full ADA compliance — at a fraction of the cost and a fraction of the wait. Built mobile-first for any dog trainer and any blind handler who refuses to spend 3 years and $70,000 waiting.