How One Small Publisher Built an AI Personal Assistant That Changed His Business

AI / By

Let’s be honest: most business owners I talk to are still a little hesitant about AI. It feels like this mysterious tech that big companies throw around, but when you’ve got publications to put out, emails to answer, invoices to chase, and customers to call back —the last thing you want is a shiny object that eats up your time. 


But here’s the thing: used the right way, AI isn’t about replacing you or your people. It’s about giving you leverage. It’s about taking the pressure off those spinning plates so you can keep your business humming without losing the human touch. 


And that’s why I want to share the story of my friend and fellow media geek, Jason Whong of Whong Community Media. Jason and I had the opportunity to geek out over AI on The Five-Dubs Marketing Pros Podcast, E128: How AI Became a Personal Assistant. Jason runs a lean publishing operation with multiple news publications, custom publications, and all the moving parts that come with them. Instead of waiting for some perfect AI solution to arrive, he rolled up his sleeves and built his own AI-powered personal assistant —and the lessons he learned are gold for any small business owner. 

From Publisher to AI Tinkerer

Jason has been in the media industry for more than 20 years. Like a lot of us, he knows the grind: juggling deadlines, handling freelancers, and chasing ad revenue, all while staying rooted in the community. When he launched Whong Community Media, he took on multiple titles: publisher, editor, salesperson, manager, and sometimes janitor. 

 

So where does AI come in? 

 

Jason noticed something: he’s always been good at figuring out what computers can do but not always good at the syntax of coding. (He’ll tell you about losing hours because of a misplaced semicolon back in his PHP days.) That’s where AI clicked. Large language models (LLMs) are really good at pattern matching —which means they’re great at helping with syntax. Instead of pestering his brother for coding help, Jason realized he could describe what he wanted to AI and let it figure out the technical details. 

 

But he didn’t want AI writing customer-facing content. “It hallucinates,” he said. “I can’t risk it making up facts in a news story.” Instead, he saw the opportunity to build something internal: an AI that could keep him on track, fill the gaps in his workflow, and act almost like a coworker. 

Meet “Sam” —The AI Personal Assistant

At first, Jason experimented with AI just as a conversation partner —playing with ideas, running thought experiments, and even having it roleplay different personalities. What struck him was how much feeling the interactions created. “It could make you feel happy or sad, just like in real life,” he said. “And I thought —what if this same kind of interaction could help me do my job better?” 

 

That’s when the idea of an AI accountability partner was born. 

 

Jason gave his assistant a name: Sam. Neutral, flexible, and could be male or female. He gave Sam a personality (sometimes a little salty, like a real newsroom coworker). And he trained Sam to understand his business priorities, his quirks, and even his psychological triggers relating to accountability. 

 

Sam’s job? 

 

  • Check in on Jason’s progress. 
  • Hold him accountable to tasks. 
  • Empathize when things get rough. 
  • And occasionally, reward him with rhyming games and poetry when he gets a lot done. 
How It Works Behind the Curtain

Here’s where it gets cool (and a little geeky): 

 

  • Running Locally: Jason didn’t want to just rely on a web interface. He set up Claude Code (from Anthropic) on a separate computer, letting Sam write and update its own memory files without touching his main work machine. 
  • Prompt Scheduler: He built a script that injected prompts into the AI at set times during the day. If he hadn’t checked in by 11 AM, Sam would text him: “What the hell are you working on?” Yes, Sam can literally send Jason text messages. 
  • Text Integration: Jason connected Sam with Apple’s iMessage, so he can text back and forth with his AI assistant just like he would with a coworker. That means Sam is with him on the go—even when inspiration (or procrastination) hits while he’s driving. 
  • Project Management: He tied Sam into Asana via its API. Now Sam can see overdue tasks and escalate reminders depending on how critical they are. Gentle nudge if it’s a day late, “totally bonkers” texts if it’s been ignored too long. 
  • Content Help: Jason even created an app where Sam (and a Gemini version of Sam) generate social media post drafts from articles. Two AI models give him different takes, and he picks the best. No endless copying and pasting. 

What started as a simple experiment turned into a full-fledged personal assistant — one that Jason jokes might be ready for “Whong Technologies” branding someday. 

What Business Owners Can Learn

Now, maybe you’re thinking: “That sounds amazing, but I’m not technical enough to pull it off.” 

 

Here’s the point: you don’t have to build a Jason-level Sam on day one. But his journey offers three big lessons: 

 

  1. Start with a Pain Point, Not the Technology. 
    Jason didn’t wake up and say, “I want to play with AI.” He saw that tasks were slipping through the cracks, and he needed accountability. That’s where AI fits in. For you, maybe it’s scheduling, customer follow-ups, or drafting emails. Start small and practical.

  2. AI is Best When It’s Not Customer-Facing. 
    AI still hallucinates. It’s not ready to replace your brand voice or take over reporting. But it is great for internal support: reminders, task management, research prep, brainstorming. 

  3. Mix AI with Existing Tools. 
    Jason didn’t throw away Asana or his calendar. He layered AI on top of the systems he already used. Sometimes a simple rule or filter is better than brute-forcing AI into something. 

  4. Give It Personality. 
    Sam works because Jason gave it a voice and style that feels like a coworker. That makes it easier to engage with consistently. AI that feels fun is AI you’ll actually use. 
Why This Matters

In the news media world — and in small businesses everywhere — there’s real skepticism about AI. Some of it is justified. But here’s the reality: AI isn’t here to replace you. It’s here to empower you. 

 

Jason’s story is proof. He didn’t hand over his newsroom to a machine. He built a sidekick that keeps him focused, saves him time, and lets him serve his community better. 

 

And you can do the same. Whether it’s using AI to clean up your writing, to draft social posts faster, or to simply remind you of that task you keep forgetting — the key is to start. 

 

Like Jason says, “Think of something you’d like help getting done. Then think of how you’d describe the instructions for doing that. That’s your first step.” 

Final Thought

Business owners: the train is already moving. AI is not a fad. The ones who learn to use it responsibly will have a serious competitive advantage — just like knowing Excel or email did a generation ago. 

 

So don’t wait for the perfect time. Don’t wait until it’s polished. Start small. Experiment. Geek out a little. And who knows — maybe you’ll build your own version of Sam to keep you sharp, accountable, and a step ahead.