A dentist who stopped practising, and kept noticing who never made it through the door.

I trained as a dentist and qualified BDS. What stayed with me wasn't the clinical work, it was watching how much gets lost before a person ever reaches the chair. The ones who couldn't fill in the form. The ones who wouldn't ring because they didn't want to be spoken to in English. The ones who were told to come back and never were. None of that is a marketing problem. It's a conversation problem, and nobody was treating it like one.

So I went and learned to build the thing itself. Twelve conversational agents across six industries, two of them live in market: one cut lead qualification time by 35 percent, another automated 40 percent of inbound counselling calls. That work was conversation design delivered on a client platform. The craft transfers; the client relationships do not. What twelve agents bought me is pattern recognition: the caller who switches language mid-sentence, the person who abandons at the first form field, the moment a script stops sounding like a person. You do not learn that from a framework. The dental booking agent is mine, built to prove the craft on my own field rather than describe it.

Twelve agents going wrong in different ways taught me one thing worth keeping: an agent earns its place by being trustworthy, not clever. Confidently wrong is the failure mode that matters, because it makes a real person act on something that was never true. So the work is narrowed to one problem: reaching the people a system currently loses, with an agent that says only what it can prove.

Arunly is my studio practice. Every engagement is led directly, with specialist collaborators brought in per build. Written first, async, fixed scope, no decks, no discovery calls that go nowhere. Based in Bengaluru.