The AI Advisor Who Went Independent to Keep Her Ethics Intact

I'm Anne Cantera, based on Long Island, New York. I advise business owners on AI adoption, design conversation systems and AI agents, and I run VoiceofAI.io, a free AI bootcamp. I'm also building Agent Modus, an agentic AI product designed from the ground up with responsible design principles. My whole practice is built around one conviction: AI should be an equalizer, not a tool for big companies to justify laying people off.

I started out running restaurants, then went to school for web and graphic design over 20 years ago. I had three kids and three family businesses along the way. Eventually I came back into UX, and spent the last four years working in AI exclusively across voice, chat, UX, UI, and agents. I got into the AI space six months before ChatGPT. I started learning NLU and NLP because I had a feeling that voice and chat were going to become really important in UX. Turned out I was right. Before AI fully took over, I was deep in UX work.

The transition was less of a leap and more of a natural progression. The tools changed. The underlying problem, how humans interact with systems, stayed the same.

I got really good at building agents. Good enough that I could see exactly what I was making: systems capable of replacing actual humans at scale. And I realised I was not willing to build those things for large companies looking for an excuse to cut headcount. I had also sat through enough poorly designed enterprise AI projects to know that the way most big organisations approach this work is genuinely bad, rushed, under-resourced, and built without real consideration for the people on the other end. So I made the decision to go independent and work with small businesses instead.

AI as an equalizer: giving a small founder access to the same quality of AI system a big company would have, without the big company's agenda behind it. That is the work I actually want to do.

I got lucky in a specific way: a friend with a consultancy had a project he needed help on, and I jumped in with him. It was not the cold outreach grind that most people go through. But it was only possible because I had been building my network for years before I needed it. Cold outreach does not work well in this field. What works is going out into the world, meeting people, having real conversations, and being genuinely useful before you ever need anything from anyone. I cannot say enough about the value of building your network as early as possible.

The pipeline follows the relationships, not the other way around.

No two days are identical. Just today I was reworking my personal website, building the most important presentation of my life, and working on content, all at the same time. That kind of parallel focus is not something I could have managed easily before generative AI. What working for myself gives me is the freedom to be an expert that people actually listen to, and the freedom to build the way I think things should be built, without 20 managers who know less about AI than I do making decisions about my work. The downside is income inconsistency.

I am not getting the same amount in my account every week. But I think that is a small price compared to the benefits. When I work from home and I am in flow, I will put in 16 hours and have no complaints. That never happened in a corporate office.

Wispr Flow, Granola, and Claude Codex are where I spend most of my time. Those are serious tools. A lot of the design tools out there are not, in my opinion. They do not create production-level code. I do not vibe code. I do spec-driven development, and that distinction matters enormously. A lot of people jumping into AI-assisted development right now are going to hit a wall when they realise the difference between something that looks like it works and something that actually holds up. Understanding that distinction, and learning how to do it properly, is probably one of the most useful lessons I could pass on.

The hardest part is the inconsistency in pay and having to provide my own benefits. That is the honest answer. Nobody going independent should be surprised by it, but knowing it is coming and experiencing it month to month are two different things. The mistake I keep making is keeping myself spread too thin: not niching down, doing too much, having 12 projects running at the same time. But honestly, I enjoy working that way.

That is just how I operate. I am someone who loves to grind and will work seven days a week and be happy about it. So at a certain point you have to decide whether something is a mistake or just your style.

I am the type of person who loves change and loves things that are new and unusual. So for me, leaving a traditional path was not scary. But I am aware that is not everyone. What I would say is: do a real personal inventory before you make the jump. Are you cut out for working 70 hours a week? Do you have financial cover, even partial, while you build? Do you know yourself well enough to stay motivated without external structure? If you are young, do not have kids, and your rent is manageable, the only way you will know is by going for it.

I do not operate on fear, so I might not be the most conservative person to ask. But I also think: what jobs are actually secure these days anyway? My general advice is go for it, but know yourself first.

Three things. Have a financial plan: know how long your runway is and what you need to cover the basics while you build. Have a good lawyer: get the contracts and structure right from the beginning, not after something goes wrong. And have a mentor you can talk to regularly, not just occasionally, someone who has been through it and will tell you the truth. Beyond that, know yourself. Know whether you are willing to be fully committed to this thing for long stretches, because that is what it actually takes.