AI Automation: What Real People are Building With AI
A farmer. A 13-year-old. A doctor. Solving real problems. That’s the real story behind AI automation.
You can’t turn a corner without seeing some media headline proclaiming 2026 is the Year of Agentic AI. Behind each of these success stories, an enterprise tech company has rolled out another solution that seemingly replaces humans. Then there are the endless creators showing you how to automate your entire content stack. Not terrible — but how many content solutions do you really need?
But here’s what slowly happening. Real people are turning to automation to build real solutions.
A farmer in Spain kept up with new national regulations by filing invoices through a voice bot on his phone. A 13-year-old in Maine built a chatbot to protect fish habitats. A doctor at the VA now gets to make eye contact with his patient instead of typing notes.
I recently entered my first automation contest, the Make.com Community Challenge (and placed as a top-five finalist on my first try.) It was exhilarating, but it also got me wondering what else people were building. What I found wasn’t robots replacing humans. It was humans using automation to analyze information faster, make better decisions, and go from idea to execution at a speed that used to require days.
That’s what this post is. It’s a tribute to the people saying, “I want to fix this. How can I do that?” It’s a celebration of what AI can do when smart, passionate people drive. We aren’t sharing THAT story enough.
What Are People Automating With AI?
The Farmer Who Replaced His Accountant With a Telegram Bot
Eduardo Cifre Sanchez — Valencia, Spain
Spanish farmers were being legally forced into electronic invoicing by nationwide mandate. Most had no idea how to comply. And hiring an accountant to help them figure it out? Not in the farm budget.
Eduardo built a voice-powered AI agent on Telegram that walks farmers through the invoicing process conversationally — in their language, on a platform they already use.
The build: Make AI Agents + Telegram + voice AI.
The result: Dozens of small farming businesses became compliant with new regulations while accountant fees went the other direction entirely. Make’s VP of Market Strategy said it was the story that stuck most from all of 2025.
Winner announcement · Make year-end feature
A Website Deployed in Two Minutes Instead of Four Weeks
Neill Human — South Africa
Web development traditionally takes weeks and costs thousands. And when a site finally goes live, leads still end up getting stuck in an inbox for days — research shows conversion drops 50% for every hour without a response.
Neill built a system where you tell the AI “build a website for a Dublin bakery” and it fetches brand configs, generates HTML, and deploys to WordPress — instantly. Every site includes an AI contact form that responds to leads 24/7 using that client’s specific configuration.
The build: Claude + Make MCP + WordPress. He connected Make’s “scenarios” with MCP tools so Claude could grab and execute. Two agents working together to build out the website.
The result: From four weeks to less than two minutes. Judges called it the best MCP demonstration in the entire competition.
Winner announcement · Watch it in action
The Autonomous AI Assistant Running an Entire Business Operation
Marco Handke — Make.com Community Challenge Winner, June 2026
Most AI workflows respond to inputs. Someone asks, the system answers. That still requires a human to initiate every interaction — which means automation can still get hung up in bottlenecks.
Marco built Mary — an autonomous AI assistant that gets to work with a single prompt. She initiates an entire work flow on her own: scheduling appointments, creating benchmarks, executing across operational workflows without waiting to be asked. (But with guardrails built in.)
The build: Make AI Agents (new architecture). Full agentic workflow with autonomous decision-making at multiple steps.
The result: Make’s announcement said his entry “didn’t just meet the brief — it redefined what an intelligent workflow could look like.”
The Chatbot That Knows Your Customer
Kai Huxmann, CEO BrainFactor GmbH — Munich, Germany
Every AI chatbot has the same flaw: it can easily answer FAQs but it can’t verify who’s asking — which means it doesn’t always understand the nuance behind the question.
Kai built a three-agent system that verifies users, retrieves their real customer data, and handles complex queries end to end — no human required.
The build: Voiceflow + Make + Airtable. Three agents working in sequence: verify, retrieve, resolve.
The result: Clients told Kai it would have taken months and an upper five-digit agency bill to build traditionally. This entry won Make’s Voiceflow community challenge.
The Ask a Farmer Knowledge Engine
Monica Poling — Make.com Top 5 Finalist, 2026
New and struggling farmers have limited access to reliable information — especially in rural communities where finding another farmer to share ideas can be hard to track down.
I built a knowledge engine that makes farm info accessible and conversational, without requiring anyone to know how to prompt an AI or navigate a complex system. Ask a Farmer also lets people buy eggs from Howling Hound Farms by chat, and throws some well-deserved love at the W.E. Heart Local tourism initiative, which does an amazing job of supporting local businesses.
Full disclosure: I built this for my sister and brother-in-law’s farm Howling Hound Farms, located in Harrow, Ontario. Their institutional knowledge about farming — the kind that takes decades to accumulate — was sitting entirely in their heads. So I put it in a Google Sheet.
The build: Make (Maia Agent) + Claude MCP + Google Sheets.
The result: The knowledge base was a Top 5 finalist in Make’s global community challenge.
→ Make.com Finalists Announcement
The Sign Language Translator
SignifyAI — ASSIST Software’s Best Innovative Minds Competition, 2025
Romania has 100 sign language interpreters for 20,000 deaf individuals. That’s a 1:200 ratio. A serious labor shortage.
SignifyAI translates spoken Romanian into real-time Romanian Sign Language animations via a photorealistic avatar — on-device, no cloud required, with an expandable vocabulary built through an automated video pipeline.
The build: Unreal Engine 5.7 rendering, on-device AI processing, automated vocabulary expansion pipeline.
The result: Sign language accessibility that doesn’t depend on scarce human interpreters. Communication access as infrastructure, not a luxury.
View Assist Innovative Minds press release.
The VA Doctor Who Finally Looked His Patient in the Eye
U.S. Department of Veterans Affairs — launched October 2025
VA doctors were spending so much time typing clinical notes that veterans said they felt like they were talking to someone’s back. That’s not care. That’s documentation theater.
The VA deployed an ambient AI scribe that listens to appointments and automatically generates clinical notes. Doctors stopped typing. Veterans got eye contact.
“My provider would look me in the eye and have an actual conversation.” — veteran survey respondent
The build: Ambient AI audio transcription integrated with VA clinical documentation systems. Piloted with more than 800,000 veterans over six months. Expanding to all VA medical centers in 2026.
The result: 71% of veterans surveyed had no concerns about the technology. Providers got hours back per day. Care felt like care again.
→ VA News
The Email Archive That Became a Directory
Monica Poling — Enchanted Artists, 2026
New Mexico is quietly home to one of the most vibrant book arts communities in the country. Two guilds – one in Albuquerque and one in Santa Fe – anchor the scene. Dozens of artists across binding, bookmaking, and paper arts. And essentially no way to find them collectively, unless you already knew where to look.
The data existed. It was just buried. I sent Claude to read more than 100 newsletters sent from the two guild organizations that were just sitting in a Gmail folder. Claude built a directory, identified the artists, extracted their details, and built a searchable data base.
The build: 100 archived newsletters searched by Claude → artist data extracted → simple HTML front end → self-updating Zapier agent handling new submissions automatically.
The result: 68 artists across two guilds, discoverable for the first time at enchanted.dataontap.dev. Proof you can do WAY more with your member archive, including turning it into a visibility engine and your biggest sales asset. Most people just don’t realize AI can do the hard work.
➤ Read More: Are you protecting your data? Or are you just hoping no one will steal your content?
What’s Your Sizzle?
Everyone has an unfair advantage. Does AI understand yours?
Enter your URL. Get your five-page Sizzle Scan analysis.
Before You Say “I Could Never Do This”
Meet Morgan Bergstrom. She’s 13. She lives in Maine. She built an AI chatbot named Cully that helps landowners and road crews identify fish-unfriendly culverts — the kind that cause flooding, road washouts, and blocked salmon migration — and guides them to replacement options and funding sources. She validated the AI’s accuracy with a decade-long conservation expert. She won first place in the National 4-H AI Challenge.
She’s not alone. Summer Wayne, a 4-H’er from Florida, built an AI project to protect green spaces in her community. Winston Fan from Iowa built an AI-powered mental health assessment system — as a high schooler. Henry Zou, also from Iowa, built an AI detection system for corn disease.
These are kids. Building real things. Solving real problems. With the same tools you have sitting in your browser right now.
So the question isn’t whether you have enough technical background.
The question is: what problem is causing friction in your business?
Automation isn’t a technology decision. It’s a knowledge decision. You already have the hard part.
– Monica Poling
Five Ideas You Can Deploy Today
A Scenario Planner for Decisions That Keep You Up at Night
Big decisions — pricing changes, new service lines, market pivots — often get made on gut feel and incomplete information. Most leaders know they’re missing something. They just don’t know what.
Build a scenario planner that models different outcomes based on your variables: pricing shifts, market conditions, unforeseen circumstances. The goal isn’t just to find the right answer. It’s to uncover the questions you didn’t know to ask.
The case: IBM’s 2026 CEO Study found leaders using AI for decision support make better calls faster — not because AI finds the answers, but because it surfaces the questions they missed.
The Knowledge Base Your Clients Pay to Access
Consultants, coaches, and trainers with 10+ years of client work
Everything you know lives in your head. You give it away in conversations, on calls, in emails — but every time you repeat yourself, you feel your eyes start to roll. Instead, take your frameworks, talks, intake forms, and methodology. Attach them to a chatbot. Give your paying customers 24/7 access to your best thinking.
The build: Google Sheets + a Make.com scenario + chatbot output. Chatbase works for a customer service-style interface. For a fully embedded experience like Ask the Farmer, build an interface from scratch.
The Report That Used to Ruin Mondays
Small business owners, agency owners, consultants with recurring clients
Reporting is painful in a way that’s hard to explain to people who don’t do it. Finding all the files. Organizing the data. Converting formats. Chasing down numbers that should be in one place. But never are
AI doesn’t just speed this up. It changes the whole relationship with the process. Pull your data sources, define your format once, and let the automation build the report while you do literally anything else.
The build: Claude + your existing data sources + a scheduled automation. At the advanced level: a Make or Zapier scenario that pulls from multiple sources, formats the output, and delivers it to your inbox or your client’s — on whatever cadence you set.
The Proposal That Writes Itself From Call Notes
Professional services firms, 5–15 employees
The gap between a great discovery call and a timely proposal is where deals go cold. Three hours of writing, formatting, pricing — by the time it lands in the prospect’s inbox, the energy from the call is already gone.
Instead, record your discovery calls. Organize your notes. You feed Claude your proposal template and your call notes, and a scoped, formatted proposal comes back in five minutes or less. Imagine how a client feels when a proposal comes back to them within the hour.
The build: Claude + your notes app of choice.
The result: Proposal time cut from three hours to twenty minutes or less. Deals that don’t go cold while you’re still writing.
AI Automation Is Yours For the Asking
AI automation isn’t something happening in the distant future — and it isn’t just happening inside Fortune 500 boardrooms. It’s being built right now by farmers, teenagers, solo consultants, and government agencies.
The common thread across every example in this post isn’t technical sophistication. It’s problem clarity. Every person here knew what they were trying to fix. AI gave them a faster way to fix it. If you can name the problem, you can start building the solution.
➤ Read More: For years, Claude AI lived in ChatGPT’s shadow, but that is rapidly changing.
Claude AI is having a moment. Read why everyone is talking about Claude AI.
TL;DR: How Real Businesses Are Using AI Automation
AI automation isn’t something that’s happening in the distant future — and it isn’t just happening inside Fortune 500 boardrooms. It’s being built right now by farmers, teenagers, solo consultants, and government agencies. The common thread across every example in this post isn’t technical sophistication. It’s problem clarity. Every person here knew what they were trying to fix. AI gave them a faster way to fix it. If you can name the problem, you can start building the solution.
Key Takeaways
- Automation is about strategy, not technology. Every builder here started with a clear problem — then figured out the tech stack.
- Your archive is already an asset. 200 newsletters became a searchable directory. What’s sitting in your inbox or Google Drive right now or on a post note or in your info dungeon right now?
- The entry point is lower than you think. Open Claude, describe your problem, ask how to put it on autopilot. That’s the starting line.
- Real results don’t require advanced builds. A discovery call fed into Claude becomes a proposal in twenty minutes. Start where you are.
- The most powerful AI applications work for people who don’t think about AI. Eduardo’s farmers. The VA’s veterans. Morgan’s fish. The goal isn’t to get everyone on board the AI story — it’s to create useful solutions for people in need.
- You already have what you need. The knowledge in your head — your expertise, your frameworks, your repeatable answers — is the asset. AI is just how you finally put it to work
Work with Me
Ask the
Expert
Turn your organizational knowledge into your biggest business asset
Most businesses have incredible expertise — but that knowledge is buried in emails, documents, and people’s heads — their website barely scratches the surface. That gap quietly costs leads, sales, and trust. Ask the Expert turns your expertise into a 24/7 AI-powered search engine and knowledge base.
Data on Tap
AI-Generated Biz Tools
Imagine how AI can visualize your expertise (and collect leads)
Data on Tap is a growing collection of AI-powered intelligence visualization tools that turn your knowledge into an interactive playground — data visualizations, calculators, and more. Using AI, you can build most of these tools in an afternoon. Follow along as I show you how. (Or contact me to build one for you.)
AI Workshops for
Organizations & Teams
Practical AI for better decision making.
AI is reshaping every industry — but most of today’s training focuses on tools. My interactive sessions start with strategy – where is your business feeling friction and how can you use AI to organize, share and monetize your knowledge. Customizable for large and small groups.
Speaking &
Keynotes
Start with your Unfair Advantage
Most businesses can’t track where their revenue actually comes from — or where it’s quietly leaking out. These keynotes and sessions show audiences how to map their revenue, close the gap between strategy and execution, and use AI as a real thinking partner along the way. You leave with frameworks you can deploy Monday morning.