From zero coding background to hardware hacker: How Cursor + a Raspberry Pi makes AI fun
Episode
28 min
Read time
2 min
Topics
Artificial Intelligence, Software Development, Psychology & Behavior
AI-Generated Summary
Key Takeaways
- ✓Hardware purchasing validation: When buying physical components for AI-assisted projects, run at least three verification passes with your AI tool before purchasing. Ask "what if this fails" and "is this still correct" iteratively. Maddie caught multiple incorrect hardware recommendations — including wrong cable types — through this triple-check process before spending money.
- ✓Brainstorm-first prompting workflow: Start any hardware or software build by dumping the entire idea into Cursor and instructing it to ask clarifying questions until all unknowns are resolved. This interview-style session produces a concrete component list, architecture plan, and purchase recommendations before writing a single line of code or buying anything.
- ✓Raspberry Pi as universal hardware bridge: A Raspberry Pi connected via Bluetooth to a thermal receipt printer, paired with a Convex database backend, creates a persistent web-to-physical messaging pipeline. The database logs every print attempt independently, providing a fallback record even when the physical printer goes offline or fails after continuous operation.
- ✓Multi-hop API chaining for legacy hardware: Connecting modern web services to outdated hardware like pagers requires accepting impractical routing. Maddie's Twitter-to-pager pipeline traverses four hops: Twitter API to Cloudflare Worker, Worker to Resend, Resend to Gmail, Gmail to the pager's legacy email address. Accepting architectural messiness unlocks hardware that would otherwise be inaccessible.
- ✓Personal API as social friction reducer: Publishing a personal API endpoint with structured data — coffee order, dietary preferences, travel dates, favorite restaurants, pet names, time zone — removes the social friction of asking friends for preferences before doing something considerate. The same endpoint becomes machine-readable for AI agents handling agentic commerce or scheduling tasks autonomously.
What It Covers
Maddie Reese, a non-coder turned hardware hacker, demonstrates three AI-assisted physical computing projects built with Cursor and a Raspberry Pi: a web-to-thermal-printer messaging system, a Twitter-connected pager, and a personal API exposing her coffee order, pet names, and San Francisco restaurant preferences.
Key Questions Answered
- •Hardware purchasing validation: When buying physical components for AI-assisted projects, run at least three verification passes with your AI tool before purchasing. Ask "what if this fails" and "is this still correct" iteratively. Maddie caught multiple incorrect hardware recommendations — including wrong cable types — through this triple-check process before spending money.
- •Brainstorm-first prompting workflow: Start any hardware or software build by dumping the entire idea into Cursor and instructing it to ask clarifying questions until all unknowns are resolved. This interview-style session produces a concrete component list, architecture plan, and purchase recommendations before writing a single line of code or buying anything.
- •Raspberry Pi as universal hardware bridge: A Raspberry Pi connected via Bluetooth to a thermal receipt printer, paired with a Convex database backend, creates a persistent web-to-physical messaging pipeline. The database logs every print attempt independently, providing a fallback record even when the physical printer goes offline or fails after continuous operation.
- •Multi-hop API chaining for legacy hardware: Connecting modern web services to outdated hardware like pagers requires accepting impractical routing. Maddie's Twitter-to-pager pipeline traverses four hops: Twitter API to Cloudflare Worker, Worker to Resend, Resend to Gmail, Gmail to the pager's legacy email address. Accepting architectural messiness unlocks hardware that would otherwise be inaccessible.
- •Personal API as social friction reducer: Publishing a personal API endpoint with structured data — coffee order, dietary preferences, travel dates, favorite restaurants, pet names, time zone — removes the social friction of asking friends for preferences before doing something considerate. The same endpoint becomes machine-readable for AI agents handling agentic commerce or scheduling tasks autonomously.
Notable Moment
Maddie's thermal receipt printer ran continuously from late October until it physically broke down — a failure she only caught because her Convex database had been silently logging every print attempt in parallel, preserving every message sent during the outage without any data loss.
Episode Transcript
I've always felt jealous of people who could coach. And then when I finally was able to kind of join that world, I just started thinking, what else was inaccessible to me that's completely accessible now? Now? And hardware was one of those things. Can you pull up your print me a message? Here is the web page where people can send me a message. And just go to mattiedreese.com/message, and you fill in your name. You fill in whatever you wanna send, and then print the ticket, and it prints out on my desk. I love it. So we're gonna test it. I have your message, Maddie, website up. I'm gonna click print ticket. There you go. Oh my god. It's so fast. Now walk us through how we might build something like this. Welcome back to How I AI. I'm Claire Vaux, product leader and AI obsessive here on a mission to help you build better with these new tools. Today, I have a really fun episode with Maddie Reese, who's building the hardware projects of my dreams using some retro gear that you might not expect. If you've ever wondered what's my friend's coffee order and then thought it would be amazing if they had an API that told me, this is the episode for you. Let's get to it. Quick word from today's sponsor, FireCrawl. If you're building with AI agents, you've probably hit the same wall. Your agent needs data from the web, but the right pages are difficult to find, buried in JavaScript, or blocked behind logins. Firecrawl is a web data API that lets agents search, scrape, and interact with the web at scale and get that clean, structured data they can actually use. Over a million developers, including myself, build on it. It's open source, and it's free to start. Stop fighting the web for data and start powering your AI agents and apps with Firecrawl at firecrawl.dev. Use code how I a I to get 10,000 free credits today. Maddie, welcome to How I a I. I was so excited when you agreed to hop on the podcast because I think you do some of the funnest projects in AI that I've seen lately. And what I love about what you're doing is you're intersecting software builds with hardware builds in a way I personally have felt has been really inaccessible until about the last six months. So what brought you to all these fun hardware projects? Kind of my journey was that I always felt jealous of people who could code. And then when I finally was able to kind of join that world, I just started thinking, oh, well, well, what else was inaccessible to me that's completely accessible now? And hardware was one of those things. So I just got really, really into it. It started with my printer and it kind of snowballed from there into a bunch of weird things. Well, I I wanna pause about …
Get the full transcript (5,148 words) + summary by email — free
One-time email with the complete transcript and AI summary of this episode. No account needed.
One email, no spam. We’ll also show you what SignalCast does.
You just read a 3-minute summary of a 25-minute episode.
Get How I AI summarized like this every Monday — plus up to 2 more podcasts, free.
Pick Your Podcasts — FreeKeep Reading
More from How I AI
Build your own company brain: the enterprise AI playbook from Stripe’s engineering team | Sharadh Krishnamurthy
Sep 7 · 50 min
Hard Fork
OpenAI Models Go Rogue + Kimi K3 Freakout + A.I. Superforecasting
Jul 24
More from How I AI
GPT-6 Astra is a banger - here’s everything I’ve built
Sep 3 · 32 min
No Priors: Artificial Intelligence | Technology | Startups
How Nuclear Will Unlock Energy Abundance with Valar Atomics Founder Isaiah Taylor
Jul 2
Books, tools, and gear mentioned in this episode
SignalCast may earn commission on purchases via these links. As an Amazon Associate, SignalCast earns from qualifying purchases.
Tools
- ConvexRecommended
“A Raspberry Pi connected via Bluetooth to a thermal receipt printer, paired with a Convex database backend, creates a persistent web-to-physical messaging pipeline”
“SPONSORS: Customer.io, https://customer.io/howiai”
“Maddie's Twitter-to-pager pipeline traverses four hops: Twitter API to Cloudflare Worker, Worker to Resend, Resend to Gmail, Gmail to the pager's legacy email address”
“Maddie's Twitter-to-pager pipeline traverses four hops: Twitter API to Cloudflare Worker, Worker to Resend, Resend to Gmail, Gmail to the pager's legacy email address”
“Maddie's Twitter-to-pager pipeline traverses four hops: Twitter API to Cloudflare Worker, Worker to Resend, Resend to Gmail, Gmail to the pager's legacy email address”
- CursorRecommended
“Maddie Reese, a non-coder turned hardware hacker, demonstrates three AI-assisted physical computing projects built with Cursor and a Raspberry Pi”
“Maddie's Twitter-to-pager pipeline traverses four hops: Twitter API to Cloudflare Worker, Worker to Resend, Resend to Gmail, Gmail to the pager's legacy email address”
“SPONSORS: Firecrawl, https://firecrawl.dev”
Gear
- Raspberry PiRecommended
“demonstrates three AI-assisted physical computing projects built with Cursor and a Raspberry Pi: a web-to-thermal-printer messaging system, a Twitter-connected pager, and a personal API”
- Thermal Receipt PrinterRecommended
“A Raspberry Pi connected via Bluetooth to a thermal receipt printer, paired with a Convex database backend, creates a persistent web-to-physical messaging pipeline”
More from How I AI
We summarize every new episode. Want them in your inbox?
Build your own company brain: the enterprise AI playbook from Stripe’s engineering team | Sharadh Krishnamurthy
GPT-6 Astra is a banger - here’s everything I’ve built
Grok Bot vs. OpenClaw: How I replaced my entire agent stack
How I turned Claude into a self-improving PM assistant | Daniel Blum (PM, Melio)
I spent $20,000 on Devin in a month. Here’s what I learned | Ryan Carson (solo founder)
Similar Episodes
Related episodes from other podcasts
Hard Fork
Jul 24
OpenAI Models Go Rogue + Kimi K3 Freakout + A.I. Superforecasting
No Priors: Artificial Intelligence | Technology | Startups
Jul 2
How Nuclear Will Unlock Energy Abundance with Valar Atomics Founder Isaiah Taylor
All-In with Chamath, Jason, Sacks & Friedberg
Jun 26
Socialists Sweep NYC, China Catches Up in Coding, AI Memory Crunch, Micron's Blowout Quarter
Odd Lots
Jun 20
How Substack Creators Are Covering This Strange Markets Era
Stuff You Should Know
Jun 19
How Chaos Theory Changed the Universe
Explore Related Topics
This podcast is featured in Best AI Podcasts (2026) — ranked and reviewed with AI summaries.
Read this week's AI & Machine Learning Podcast Insights — cross-podcast analysis updated weekly.
You're clearly into How I AI.
Every Monday, we deliver AI summaries of the latest episodes from How I AI and 192+ other podcasts. Free for one show.
Start My Monday DigestNo credit card · Unsubscribe anytime