AI agents that do real work every day

An AI agent is a program that gets a task, reads what it needs, decides what to do and does it. Besides answering, it acts: it sorts an email, updates a record, alerts someone or drafts a reply.

Unlike a chatbot, it works inside your processes instead of just chatting. Unlike classic automation, it understands text, nuance and cases nobody wrote down in advance.

It has limits too. It won't handle everything on its own. You don't install it on Friday and have it running on Monday. And it won't decide for you how your company should work.

In sales, it reads every contact that comes in, judges whether it fits and assigns it to the right person. Nobody has to hand them out manually at the end of the day.

In customer service, it answers the questions that keep coming back, by phone, by voice on your website or by text. What it doesn't know, it hands to a person with the context already summarised. If it talks to your customers, it has to be accessible to anyone.

In marketing, it watches ad performance, flags when something goes wrong and adapts content to several languages without losing the tone. In operations, it checks every day that what should work does work. In management, it answers plain-language questions about your own data, without waiting for someone to pull a report.

Since April 2026 I have been Head of AI GTM Engineering at Sesame HR, with a mandate over the internal AI projects of the whole company. There I have more than ten tools in production across five markets. They follow competitors, answer questions about business data, prepare content in several languages, watch ad performance and check the website end to end.

Before that I was Global Lead, Automation & Tech at Dentsu International (Dentsu X and iProspect), in London. That is where I built GeoAdsFinder, which ended up being used across the network. I have worked in agencies and large companies, in marketing, revenue and product.

Outside Sesame HR I build AIsWise: voice and text AI assistants with accessibility built in to WCAG 2.2, the international web accessibility standard. In 2021 I co-founded Chat Ergo Bot, a conversational AI startup I led, before ChatGPT existed. I have spent years seeing what works once the demo is over.

A process that already exists. If nobody knows today how a task gets done, an agent won't guess it. It will automate the mess.

Data it can read. If the information lives in one person's head or is spread across ten spreadsheets, that is the first job.

A person accountable for the result, who decides what is right and what needs fixing. And a record of what the agent does: what it read, what it decided and why. With that, anyone on your team can check its work.

A demo is built to go well: clean data, an ideal case and nobody asking what happens when it gets something wrong. In a real company the data is incomplete and the odd cases are half the job. And someone has to answer when something goes wrong.

What is usually missing is the system around it: where the data comes from, who supervises and how you measure whether the agent helps or gets in the way.

First I learn how your company works: who does what, with which tools and where things get stuck. Sometimes the answer is an agent. Other times it is something simpler, like sorting out where information lives. I pick what works.

When it's time to build, I start with a single task. I check that it helps and leave it running before moving to the next one. Your team stops copying data from one place to another and spends that time on customers. And every decision the agent makes is on record.