In this 47-minute conversation on the Tech Effect podcast by TestDevLab, I discuss why AI implementation isn’t just for big tech companies — it’s for every business, regardless of size or technical sophistication. We cover data literacy as a foundational skill, how to identify where AI can actually help in your operations, and why the real barrier to adoption is psychological, not technical.
Josh from TestDevLab invited me on to talk about my journey from architecture to AI, what I’ve learned building an education platform for over 100,000 people, and the practical reality of AI implementation for businesses that don’t have engineering teams. This one goes deep — 47 minutes covering everything from parametric design in Danish robot labs to why your 70-year-old community member can build no-code AI solutions.
Listen to the Full Episode
Also available on Spotify, Apple Podcasts, and Amazon Music.
Timestamps
| Time | Section |
|---|---|
| 0:00 | Coming Up |
| 1:00 | Intro |
| 2:12 | Goda’s Journey |
| 9:14 | Translating AI So Anyone Can Understand It |
| 15:26 | Barriers to AI Implementation |
| 18:44 | How to Identify Areas for AI Implementation |
| 26:26 | The Importance of User Feedback |
| 31:20 | Data and Its Importance |
| 38:06 | What Skills Business Leaders Need for AI Adoption |
| 45:35 | Follow Goda Go’s Journey |
Key Discussion Points
From Architecture to AI
I started in architecture in Lithuania, got a scholarship to a top-10 European university in Denmark, and spent my days in robot labs with CNC machines and robotic arms. That obsession with tech and data-driven design eventually pulled me out of architecture entirely — first into big data analytics in the travel industry, then into AI.
What most people don’t realize about architecture is that it’s deeply technical. You’re working with parametric design, computational modeling, and systems that have to account for structural loads, light, airflow, and human behavior all at once. That kind of thinking — holding multiple complex variables in your head and designing systems that balance them — is exactly the skill set you need for AI implementation. I didn’t know it at the time, but architecture was the best possible training ground for what I do now.
Translating AI for Everyone
The biggest problem isn’t the technology. It’s translation. Most AI content is written by engineers for engineers. What I do is bridge that gap — take complex concepts and make them accessible so that a marketing director, a small business owner, or a 70-year-old entrepreneur can actually implement this stuff.
I’ve seen this gap firsthand in my community. People who are brilliant at what they do — running businesses, managing teams, creating products — feel locked out of AI because the conversation is dominated by technical jargon. Terms like “fine-tuning,” “RAG,” “embeddings,” and “tokens” create an artificial barrier. The reality is that most people don’t need to understand the engineering behind AI. They need to understand what it can do for their specific situation, in plain language, with a clear next step.
That’s what I mean by translation. Not dumbing things down — making them actionable for people who have real work to do.
Barriers to AI Implementation
The real barriers aren’t technical. They’re psychological. People think AI is “not for them” or that they need to be technical. They don’t. The tools are accessible. The no-code builders exist. The real barrier is just getting started.
In the podcast, Josh and I go deeper into the specific barriers I’ve observed. Here are the most common ones:
- The “I’m not technical” belief. This is the single biggest barrier. People disqualify themselves before they even try. The truth is that tools like ChatGPT, Claude, and no-code platforms have made AI accessible to anyone who can write a clear sentence.
- Information overload. There are so many AI tools, so many opinions, so many “you MUST use this” posts that people freeze. They don’t know where to start, so they don’t start at all.
- Fear of looking foolish. Nobody wants to be the person in the room who asks “what’s a prompt?” So they stay quiet, pretend they understand, and never actually implement anything.
- No clear use case. “AI can do anything” is actually terrible advice. When everything is possible, nothing feels actionable. People need specific, concrete starting points.
How to Identify Areas for AI Implementation
This was one of the meatiest parts of our conversation. Josh asked me how a business that’s not in tech should even begin thinking about AI. My answer: start with the pain, not the technology.
Here’s the framework I use with businesses and community members:
- List your recurring tasks. What do you or your team do every day or every week that feels repetitive? Email sorting, data entry, report generation, customer response drafting, social media scheduling — these are all candidates.
- Identify the bottlenecks. Where does work get stuck? Where do things take longer than they should? That’s usually where AI can have the most immediate impact.
- Score by effort vs. impact. Not every AI implementation is worth it. Focus on tasks where the effort to set up automation is low but the time saved is high.
- Start with one workflow. Don’t try to transform everything at once. Pick one workflow, implement an AI solution, measure the results, and then expand.
This is fundamentally the same approach I use when building AI Second Brain — start with a real problem, build a system around it, then iterate.
The Importance of User Feedback
One thing I emphasized in the conversation is that AI implementation is never “set it and forget it.” The tools are powerful, but they need feedback loops. If you deploy an AI chatbot for customer service and never check what it’s actually saying, you’ll have problems. If you use AI to draft emails and never review the output, your voice will drift into generic AI-speak.
The best implementations I’ve seen treat AI as a draft machine, not a finished-product machine. AI creates the first version. A human reviews, adjusts, and approves. Over time, the AI gets better because you’re feeding it better context and corrections. But that feedback loop is essential — without it, quality degrades.
Data Is Everything
Once you work with data, you start seeing everything through data. Every business generates data — the question is whether you’re capturing it and using it. AI without good data is just guessing. The companies that will win aren’t the ones with the fanciest models. They’re the ones with the best data.
This is a point I’m particularly passionate about. There’s a widespread misconception that AI value comes from the model — that whoever has the most advanced AI wins. That’s not how it works in practice. The model is a commodity that’s rapidly getting cheaper. Your data, on the other hand, is unique. Your customer interactions, your sales patterns, your operational workflows — that data is yours and nobody else has it. The real competitive advantage is capturing that data, structuring it, and using AI to extract insights from it.
Why Data Literacy Matters for Everyone
You don’t need to be a data scientist to be data literate. Data literacy means understanding what data your business generates, knowing which metrics actually matter, and being able to ask meaningful questions of that data. It means knowing the difference between a vanity metric and an actionable insight.
Here’s what basic data literacy looks like in practice:
| Data Literacy Skill | What It Looks Like in Practice |
|---|---|
| Knowing what you measure | Being able to list the 5-10 key metrics for your business |
| Understanding data sources | Knowing where your data comes from and how reliable it is |
| Asking the right questions | “Why did conversions drop last week?” vs. “What are our total views?” |
| Recognizing patterns | Spotting trends and anomalies without needing a dashboard |
| Separating signal from noise | Ignoring vanity metrics and focusing on what drives outcomes |
This isn’t advanced analytics. This is the basic skill set that every business leader — and honestly, every professional — needs to develop. AI amplifies data literacy. It can surface patterns you’d never see manually. But it can only do that if you know what to look for and what questions to ask.
AI Implementation for Small Business
You don’t need to be a Fortune 500 to benefit from AI. A small business with one well-implemented AI workflow can gain a competitive edge that used to require an entire department. Start with the problem, not the technology. Find the bottleneck, then ask: can AI help here?
Some concrete examples we discussed:
- A local restaurant using AI to analyze customer reviews and automatically surface common complaints or requests — no data team required, just feed reviews into a simple AI prompt.
- A freelance consultant using AI to draft proposals based on templates and past successful bids — cutting proposal time from 4 hours to 30 minutes.
- A small e-commerce shop using AI to generate product descriptions, write email campaigns, and analyze which products get the most returns and why.
- A real estate agent using AI to draft property listings, summarize market data, and automate follow-up emails to leads.
None of these require custom AI models or engineering teams. They require a clear problem, a basic tool (often just ChatGPT or Claude), and the willingness to experiment. That’s it.
What Skills Business Leaders Need for AI Adoption
Toward the end of the episode, Josh asked what skills business leaders should develop. My answer wasn’t “learn Python” or “take a machine learning course.” It was this:
- Learn to write clear instructions. If you can write a good brief for a freelancer, you can write a good prompt for AI. The skill is the same: defining the goal, providing context, and specifying the output format.
- Develop systems thinking. Understand how your business processes connect. AI is most powerful when it’s integrated into a workflow, not used as an isolated tool. If you can map your processes, you can identify where AI fits.
- Get comfortable with imperfection. AI outputs aren’t perfect. They’re first drafts. Leaders who wait for 100% accuracy before adopting will wait forever. The ones who treat AI as a “good enough to start” tool and iterate from there will move fastest.
- Build feedback loops. Check what your AI systems produce. Correct them. Refine the prompts. This is where the real value compounds over time.
- Stay curious, not anxious. The pace of change is fast, but you don’t need to learn everything. You need to learn what matters for your situation. And that’s a much smaller, more manageable set of knowledge. If you want to see this curiosity-driven approach in action, check out my TEDx talk on being a better early adopter.
The Practical Playbook: Getting Started with AI Implementation
If you listened to the full episode and want a concrete action plan, here’s the playbook I recommend for any business, regardless of size:
Week 1: Audit. Spend one week documenting every repetitive task you or your team performs. Don’t judge or filter. Just list everything.
Week 2: Prioritize. Score each task on two axes: how often it happens and how much time it takes. The tasks that are frequent AND time-consuming are your first AI candidates.
Week 3: Experiment. Pick the top task from your list. Try using AI (ChatGPT, Claude, or a no-code tool) to assist with it. Don’t aim for full automation — aim for assistance. Can AI draft the first version? Can it organize the data? Can it suggest the structure?
Week 4: Evaluate and Expand. Did the experiment save time? Was the quality acceptable? If yes, refine the workflow and make it repeatable. Then move to the next task on your list. If you want to build this into a sustainable system, building your own AI bot is a great next step for automating recurring interactions.
That’s it. Four weeks, one workflow at a time. No massive investment, no technical team, no disruption. Just steady, practical progress.
About the Podcast
Tech Effect by TestDevLab — interviewing industry leaders and experts on the topics and trends shaping the tech industry. Available on YouTube, Spotify, Apple Podcasts, and Amazon Music.
Links
Frequently Asked Questions
How can a non-technical business start implementing AI?
Start with one repetitive, time-consuming task — like drafting emails, sorting customer feedback, or generating reports. Use an accessible tool like ChatGPT or Claude to assist with that specific task. Don’t aim for full automation. Aim for a useful first draft that you review and refine. Once that workflow saves consistent time, move to the next task.
What is data literacy and why does it matter for AI?
Data literacy is the ability to understand what data your business generates, which metrics actually matter, and how to ask meaningful questions of that data. It matters because AI is only as good as the data you feed it. A business leader who understands their data can use AI to surface patterns and insights that would take a team of analysts to find manually.
Do I need to learn to code to use AI in my business?
No. The vast majority of valuable AI implementations for small and medium businesses require zero coding. No-code tools, conversational AI assistants, and pre-built integrations cover most use cases. The skill you need is the ability to write clear instructions and understand your business processes — not programming.
What’s the biggest mistake businesses make with AI implementation?
Trying to do too much at once. Businesses that attempt to “AI-transform” everything simultaneously usually end up with a dozen half-finished projects and no measurable results. The best approach is to pick one workflow, implement AI assistance, prove the value, and then expand. Slow and steady beats ambitious and abandoned every time.
How does Goda Go help businesses implement AI?
Through the Autonomee community, YouTube tutorials, and direct education. I focus on making AI accessible for non-technical professionals — translating complex concepts into plain language and actionable steps. My approach starts with real business problems, not technology hype, and emphasizes building sustainable systems rather than chasing every new tool release.
