David is an AI Integrator at Trio Media, bringing the power of AI into our daily work to streamline and automate wherever possible, from simple automations to custom AI agents, chatbots and more.
David keeps us on top of the fast-moving AI world, making sure we’re up to date with the latest trends, models and everything new happening in the AI space.
What drew you to working in AI and web development?
From a young age, I was interested in technology and everything related to computers – from programming games to 3D modelling, websites and beyond.
I am a creative person, so combining my passion for technology with creativity naturally led me down the path of development. I have explored many areas within IT since a young age, but when I found out that you can save a website’s source code as a file and then mess around with it, I knew this was going to be my thing.
As time went on, I got into more advanced work, and my career choices steered me towards web development. Over the years, I took on bigger builds where logic and problem-solving were the main focus rather than design and visuals.
With the arrival of AI and its assistance in programming, I realised that writing code is no longer an issue – AI can do it for me. So I gradually shifted from using AI to write the code to asking: what can we do if we harness AI at the end of that code? What if we give an app, website, or function some “brain” and “understanding”?
And that was my natural shift – from using AI as a tool to building it in as a feature.
Are there any key lessons you’ve learnt that influence your work today?
Many – and still counting.
The first lesson I learnt when shifting to AI development was the value of planning up front, rather than changing direction as I go. In my day-to-day work, writing code by hand is largely a thing of the past. I still orchestrate everything and act as the key architect, but the typing itself is gone. Before AI, I could carve and sculpt a project as I went along; now, far more time goes into planning before a single line is written.
Another key lesson, and one I’m still learning every day as the models improve, is understanding AI’s capabilities and how to use it correctly. The output is only as good as the input, and context matters. AI doesn’t know things you don’t tell it; it will make assumptions, and in most cases, you don’t want that. I treat the AI I use daily as my assistant: I’m descriptive, we plan together, and I always ask it to challenge my approaches and decisions.
What current AI trends do you think will have the biggest impact on how businesses will work?
I think the biggest AI trend right now is adoption. Many businesses are only just discovering how AI can help them and change the way they work. Companies that adopt AI first will be way ahead.
Another trend is happening more quietly: AI agents. Even with something as simple as ChatGPT or Claude online, you can already “do things”. There are many integrations available, and you can tell your AI model to create a document or send an email without ever leaving the chat. AI is no longer just a simple chatbot, even though most people still use it the same way they did before. But they can sense that AI is capable of more.
We will see agents becoming the new norm, just as chat did a few years ago.
What do you believe is the next big thing in AI?
Beyond agents, the next big thing is already happening: open-source AI models that can run on consumer-grade hardware.
Free, open-source models are being released every month, and they’re getting better and better. At the same time, hardware requirements are being optimised and consumer hardware itself keeps moving forward. The models you can download and run locally today are already genuinely capable of intelligent tasks – things that would have required a cloud API and a subscription not long ago.
This matters most for data privacy: for anyone working with sensitive information, running a model locally means your data never leaves your PC. I think we’ll see the gap between open-source and “big boy” models keep narrowing, and for a huge range of real-world tasks, local models will simply be good enough.
Is there anything you think people get wrong when using AI?
As I mentioned, most people are only scratching the surface. But there’s also a growing divide between people who ignore AI or use it very little in everyday life and those who dive into it fully. We’re at a point where AI’s presence can’t be ignored – if you want to be efficient, you have to use it in some shape or form.
That said, a human must always stay in the loop and remain the decision-maker. AI can do the heavy lifting on reasoning, analysis and repetitive tasks, but the judgement calls stay with us. Ultimately, it’s about finding the right balance between not using it enough and overusing it.
Do you have any tips for integrating AI strategically into their business?
Commit to it and do it right. Many people think AI will replace jobs and are afraid of bringing it into the business, but the way I see it, it can make you superhuman if you let it. Every business should have an AI adoption strategy.
The first step is probably to improve day-to-day work before diving deep into AI. Identify where most time is spent and see if any of it can be automated or AI-enabled: find the problem first, then ask whether it’s worth solving. Always have a plan for when the technology fails, and set goals so you know when it succeeds.
Want to do more with AI?
With a growing team of AI specialists, we’re here to help integrate AI into your business, whether to improve your marketing or streamline your operations. Have a question or an idea in mind? Speak to our team today.
