The last mile of AI
runs through us

Get good at putting AI to work.

Knowing what AI can do is the beginning. The useful part is choosing a real problem, building a workflow, and checking the result. That is the last mile. We want to learn it together.

Five departments.
Find where you fit.

Read what each one does, then pick a first and a second choice when you apply.

What we learn, shared.

The Commons is our open resource collection: practical workflows, guides, and resources worth your time. Start here, whether you join the club or not.

The labs solved capability. Nobody solved adoption.

Billions have gone into making these systems more capable. Almost nothing has gone into the part that decides whether any of it matters: whether a person sitting in front of a real deadline knows how to get something correct out of the machine, and knows when not to trust it.

That gap is the last mile of AI. It is not a research problem and it will not be closed by a better model. It is a practice problem, made of workflows, judgement and habits, and it is closed one person and one process at a time.

Business students are not spectators to this. Somebody inside every company is about to be handed responsibility for how these tools get used, and most of them will not be engineers. The job is coming, it is commercial rather than technical, and almost nobody is being trained for it. We work at the adoption layer.

Our ambition is plain. We want every student here using AI better than the average professional in their field, and we want the guides that get them there to be public.

Mostly teaching.
Then get to work.

Most of what we run is teaching: sessions where you leave with something that works, people who use these tools daily showing you how, and a chance to build under pressure. Our planned autumn programme is below; we will publish exact dates and locations as they are confirmed.

We want to study how adoption actually happens.

Most research asks what these models can do. Far less of it asks what happens when somebody with a deadline and no technical background sits down and tries to use one for real work. That gap is the reason this club exists, and we are well placed to look at it, because we are the people it happens to. We run it on two tracks.

Track 1 · Applied research. Research on how AI gets implemented in practice. We want faculty involved who know how research is designed. Part of it runs on ourselves: track what members actually try, where they give up, and which workflows are still in use a month later.

Track 2 · Verified knowledge. We track the AI news that matters, test tools with our Projects team, and publish what holds up in The Commons. AI safety and responsible use are a standing topic.

Curiosity first.
Then contribution.

Help choose what we explore and make the club useful to others. The application shows the questions for each department; we will explain roles and expected commitment in the chat.

We select for motivation and what you have built, not your CV. Expect a few questions for each department you pick, a 30-second to 2-minute video, and a short chat with the department head.

Meet the team.

Students bringing experience from AI infrastructure, venture capital, consulting, and automation. Two co-presidents split the club into an internal and an external side, and every department has its own head. We are building the club we wanted to find.

Meet people who can show the work.

Lead a practical session, host a company visit, sponsor an event, bring us a process our Projects department can help you automate, support our open resources, or help us build connections between members and opportunities in your field. We would like to hear from practitioners and teams in Lisbon and beyond.

Projects are pro bono, built by students in Nova's name and handed over for good.

Contact via LinkedIn