Subscribe to Hard Mode First
>600 subscribers
Hard Mode FirstChatGPT Saved My Life (No, Seriously, I’m Writing this from the ER)
How using AI as a bridge when doctors aren't available can improve patient-to-doctor communications in real time emergencies
Hard Mode FirstHow to Plan an Annual Family Summit
Simple strategies for setting goals and Priorities with Your Partner for the year ahead
Hard Mode FirstHow I Used AI to Save My Life in 77 Prompts: A Debrief
Reflecting on best practices, lessons learned, and opportunities to improve AI-assisted medical triage
Subscribe to Hard Mode First
>600 subscribers
Hard Mode FirstChatGPT Saved My Life (No, Seriously, I’m Writing this from the ER)
How using AI as a bridge when doctors aren't available can improve patient-to-doctor communications in real time emergencies
Hard Mode FirstHow to Plan an Annual Family Summit
Simple strategies for setting goals and Priorities with Your Partner for the year ahead
Hard Mode FirstHow I Used AI to Save My Life in 77 Prompts: A Debrief
Reflecting on best practices, lessons learned, and opportunities to improve AI-assisted medical triage
Share Dialog
While beautiful, for many of us, this felt like an unwelcome do-over from January’s storm, which left cold, icy mounds of snow on the street for weeks.
At a recent block meeting of The Manhattan 75, we spoke about how that first storm left many senior residents on our block stranded in their homes for longer than expected, given how narrow foot paths at intersections made it too dangerous to cross sidewalks safely. I also lamented how terrible it was to be trapped in my home with a leaky ceiling for several days, unable to shovel the roof on my own.
We decided when the next storm came around, we’d like to do something about it.
So this weekend, while the snow fell on West 75th Street, we mobilized a small group of on-call volunteers, who signed up in shifts to indicate their availability to help with small tasks (ie: running errands, shoveling street corners, or even in-home help where appropriate.)
And then we turned it into an app.
Share Dialog
While beautiful, for many of us, this felt like an unwelcome do-over from January’s storm, which left cold, icy mounds of snow on the street for weeks.
At a recent block meeting of The Manhattan 75, we spoke about how that first storm left many senior residents on our block stranded in their homes for longer than expected, given how narrow foot paths at intersections made it too dangerous to cross sidewalks safely. I also lamented how terrible it was to be trapped in my home with a leaky ceiling for several days, unable to shovel the roof on my own.
We decided when the next storm came around, we’d like to do something about it.
So this weekend, while the snow fell on West 75th Street, we mobilized a small group of on-call volunteers, who signed up in shifts to indicate their availability to help with small tasks (ie: running errands, shoveling street corners, or even in-home help where appropriate.)
And then we turned it into an app.
Storm Link dispatches volunteers on the block with on-call requests for help through automated email introductions.
Here’s the flow:
Neighbors request help through a simple form. No account needed, just their name, address, and what they need
Volunteers sign up for shifts and indicate what tasks they’re comfortable with (shoveling, errands, in-home help, etc.)
When a request comes in, the system automatically notifies all on-call volunteers via email and invites them to “claim” a new task.
They take it from there. We stay out of the way and let neighbors help neighbors directly.
On Sunday morning, we sent a request via Google Form to seed our first set of volunteers. On Monday morning, we sent the web app around to continue to recruit volunteers and notify block residents (approx. ~300 people on a mailing list) that they are welcome to submit requests.
By mid-afternoon on Monday, we had 20 volunteers signed up in shifts and successfully completed our first request for help (for a neighbor who needed some errands done so he could stay at home with his 5-year-old autistic son).
One thing I love about Storm Link is its versatility. Now that the infrastructure is in place, we’ll be ready for the next emergency too. The backend stays the same, we can just reskin the frontend for whatever comes next (ie: hurricane season, heat wave, another blizzard...) The bones are there.
This is what I’ve started calling “pop-up software,” apps built for a specific moment, a specific community, a specific need.
One thing I hear a lot from people who are new to building apps with AI is how discouraging it can be to still feel so far away from building “production-grade software.”
But this feeling is sort of like feeling like it’s not worth learning how to cook because you know you’ll never be good enough to open a Michelin Star restaurant.
Millions of people all over the world find it incredibly helpful to learn how to cook, even without aspirations to open their own restaurants one day. This is why there’s such a massive industry in everything from at-home cookbooks to reality TV shows about amateur chefs.
The reality is: If you can whip up even a few dishes for your family, any investment in learning how to cook pays off immensely. You are applying that skill in a way that makes a difference for you. Of course, the more comfortable you become in a kitchen, the more advanced meals you’ll be able to cook on your own.
Building with AI is the same thing. And the more comfortable you find yourself around a codebase, the more dishes… err, I mean apps… you’ll be able to build on your own.
Not every app needs to scale to millions of users. They don’t all need enterprise security audits. They just need to work well enough, right now, for the people who need them.
By the way… I’m looking for an excuse to put Storm Link on GitHub or fork it for another community. So if you’re interested in building a hyper-local pop-up app for your block, neighborhood, or building, give me a shout at bethany@buildfirst.ai.
P.S. This entire system was built without writing a single line of code myself. AI tools used included: Claude Code and ChatGPT. Dev tools included used Supabase, Resend, Github, and Vercel.
Storm Link dispatches volunteers on the block with on-call requests for help through automated email introductions.
Here’s the flow:
Neighbors request help through a simple form. No account needed, just their name, address, and what they need
Volunteers sign up for shifts and indicate what tasks they’re comfortable with (shoveling, errands, in-home help, etc.)
When a request comes in, the system automatically notifies all on-call volunteers via email and invites them to “claim” a new task.
They take it from there. We stay out of the way and let neighbors help neighbors directly.
On Sunday morning, we sent a request via Google Form to seed our first set of volunteers. On Monday morning, we sent the web app around to continue to recruit volunteers and notify block residents (approx. ~300 people on a mailing list) that they are welcome to submit requests.
By mid-afternoon on Monday, we had 20 volunteers signed up in shifts and successfully completed our first request for help (for a neighbor who needed some errands done so he could stay at home with his 5-year-old autistic son).
One thing I love about Storm Link is its versatility. Now that the infrastructure is in place, we’ll be ready for the next emergency too. The backend stays the same, we can just reskin the frontend for whatever comes next (ie: hurricane season, heat wave, another blizzard...) The bones are there.
This is what I’ve started calling “pop-up software,” apps built for a specific moment, a specific community, a specific need.
One thing I hear a lot from people who are new to building apps with AI is how discouraging it can be to still feel so far away from building “production-grade software.”
But this feeling is sort of like feeling like it’s not worth learning how to cook because you know you’ll never be good enough to open a Michelin Star restaurant.
Millions of people all over the world find it incredibly helpful to learn how to cook, even without aspirations to open their own restaurants one day. This is why there’s such a massive industry in everything from at-home cookbooks to reality TV shows about amateur chefs.
The reality is: If you can whip up even a few dishes for your family, any investment in learning how to cook pays off immensely. You are applying that skill in a way that makes a difference for you. Of course, the more comfortable you become in a kitchen, the more advanced meals you’ll be able to cook on your own.
Building with AI is the same thing. And the more comfortable you find yourself around a codebase, the more dishes… err, I mean apps… you’ll be able to build on your own.
Not every app needs to scale to millions of users. They don’t all need enterprise security audits. They just need to work well enough, right now, for the people who need them.
By the way… I’m looking for an excuse to put Storm Link on GitHub or fork it for another community. So if you’re interested in building a hyper-local pop-up app for your block, neighborhood, or building, give me a shout at bethany@buildfirst.ai.
P.S. This entire system was built without writing a single line of code myself. AI tools used included: Claude Code and ChatGPT. Dev tools included used Supabase, Resend, Github, and Vercel.
3 comments
I wrote a blog post about the pop-up app that I vibe coded for my block to help manage snowstorm support for our neighborhood. As it turns out, you don’t need enterprise-level infrastructure to build software that can make a difference. Not every app needs to scale to millions of users. They just need to work well enough, right now, for the people who need them. https://hardmodefirst.xyz/pop-up-software-for-snowstorm-support
scaling is mostly a trap anyway, love seeing stuff that actually helps real people
Super satisfying to really know your customers / users personally too. It's a much intimate experience overall.