UBS

Two halves, and the introduction is the smaller one. Thirty to forty seconds is plenty: who you are, what you study, and one line that actually distinguishes you, then move on, because every mark in this answer sits in the second half. That second half is more specific than most candidates notice. Areas or roles at UBS is asking you to name a division, and UBS is not mainly an investment bank. It is the world's largest wealth manager, and it runs four businesses: Global Wealth Management, Asset Management, Personal and Corporate Banking, and the Investment Bank. Name one. Saying you are interested in UBS generally, or in finance generally, wastes the question. If it is the Investment Bank, go a level finer and say Global Banking or Global Markets, and if you can, a sector or product within that. Then the why, which has to attach to something real rather than to an adjective. A deal UBS worked on, a piece of UBS research you read, something somebody said at an event you went to, or a piece of work you have done that pointed you this way. One concrete anchor beats three sentences of enthusiasm. For a spring week there is an honest route open to you that is not open at summer stage: you are allowed to be exploring. Saying you want to test whether wealth management or the investment bank suits you, and naming which one you lean towards and why, reads as self aware rather than uncommitted, provided you still commit to a direction. What does not work is the opening everybody else uses, which is a second year studying something who is passionate about finance and eager to learn.

Two questions that sound like one, and almost everybody answers the first and repeats it as the second. They are different things. Goals are what you intend to do. A definition of success is what you will measure yourself against, which is a values question wearing a career question's clothes, and it is where this is actually marked. On goals, keep it short and make it a staircase rather than a destination: this spring week, then a summer internship in a named area, then a graduate seat, then something further out that is about capability rather than title, such as knowing a sector well enough to have a view people ask for. Naming the area makes it concrete and ties back to the first question of the set. On success, two traps sit either side of the road. Saying money or titles is the obvious one. The less obvious one is the overcorrection, insisting that money does not matter at all, which nobody believes and which sounds rehearsed. The strongest answers define success as a kind of competence or contribution you can describe: being the person a team gives the difficult piece of work to, doing work that is still useful after you have left it, being trusted by a client. Then do the thing that separates a real answer from a polished one, which is to say briefly where your definition came from. One honest sentence about a person you watched, a job you did, or a moment that changed what you were aiming at carries more weight than three well constructed ones. You can nod to UBS at the end, since client centricity and sustainable impact are its own language, but say it in your words rather than reciting the principles back to them.

The first half is a warm up and the second half is the actual test, so do not spend equal time on them. Thirty seconds on sources, the rest on the topic. On sources, name two or three you genuinely read and be specific, because a list of five famous mastheads with no evidence behind it is the single most common answer and it reads as a list of famous mastheads. A newsletter, a podcast, a bank's own research. UBS publishes a great deal of research and naming a piece of it here is free, relevant and almost nobody does it. Then the topic, which you should choose before the assessment rather than on the spot. Pick something narrow enough that you can hold ninety seconds on it, and run it in four beats: what is happening, why it is happening, who gains and who loses, and what you think happens next. The fourth beat is the one most candidates skip and the one that separates following a story from having read a headline about it. Carry one number and get it right, because a single correctly remembered figure does more work than a paragraph of narrative. If you can find a topic with a UBS angle, take it, because it quietly answers why UBS at the same time: flows into wealth management, Swiss capital requirements and what the Credit Suisse integration changed, what rate moves do to net interest income, private markets arriving in wealth portfolios. Two things to avoid. Do not pick a subject so enormous, such as artificial intelligence or tariffs, that nothing you say about it could be specific. And do not pick something you read once last week, because the follow up in your own head should be whether you could still talk about it in a month.

Read the words outside of school or work, because they are doing deliberate work and the commonest failure here is ignoring them. A group coursework project is school. A placement is work. Reaching for either answers a question that was not asked, and it tells them you were running a prepared answer rather than listening. What counts is sport, music, a society, volunteering, coaching, a side business, something you organise, a part time thing nobody made you do, a caring responsibility at home. None of these is too ordinary. The more ordinary it sounds, the more of the work you have to do yourself in connecting it, and that connection is exactly what is being marked. Build it in three moves. Name the activity and your role in one sentence. Then give one specific episode rather than a general description of what you do, because a scene is evidence and a summary is a claim. Then make the link to the job explicit and concrete. Not that it taught you teamwork, which is what everybody says and means nothing. Something with an edge to it: running a society's budget meant being the person who told people no and then had to work with them again the following week, so I learned to deliver an unwelcome answer without making it personal. Choose the skill with UBS in mind. Their behaviours are accountability with integrity, collaboration and innovation, so pick one of those and prove it properly rather than claiming all three thinly. And if the activity involved money, people or a deadline that could not move, say so, because those three things are what the job is made of.

New for this cycle, barely anybody has prepared for it, and it has a trap on each side. Saying you do not really use AI reads as either untrue or incurious. Saying it did the work for you reads as somebody who cannot be handed a deliverable. You want the middle: a person who uses it hard and checks it harder. The enormous advantage here is that the question tells you its own marking scheme. It names skewed data, transparency and quality checks, and most candidates will answer only the efficiency part and never touch the other three. Address all four by name and in order. On efficiency, be specific and be modest about what it actually did. Drafting a first version you then rewrote, summarising a long document so you knew which part to read properly, writing code to clean a dataset, arguing against your own conclusion so you could find the holes. Say roughly what it saved you and, more importantly, what you spent that time on instead, because that is where judgment shows. On skewed data, say you thought about where the inputs came from and who is missing from them: a dataset covering only listed companies, survey responses only from people who bothered to reply, a model trained mostly on one market. Naming the specific gap in your own case is far better than a general statement that AI can be biased. On transparency, say who you told. Did you disclose it to a supervisor, cite it, write it into your method, check what your university's rules actually said. Saying you looked up the rule is a boring, credible detail that very few people will offer. On quality checks, which matters most, give the check itself rather than saying you verified the output. Tracing a claim back to the primary source, recomputing a figure by hand, having somebody who knows the subject read it before it went anywhere. If you ever caught it being confidently wrong, lead this section with that story, because it is the single most persuasive thing you can say. Finish with the rule you now use for when you would reach for it and when you would not.

Do not treat this as the exit door. It is marked required like every other question, the camera is still running, and saying nothing further, thank you, throws away the best two minutes in the whole assessment, because it is the only slot where you choose the subject. It is also the only question you can genuinely write in advance, since it sits at the end of every set, so draft and rehearse it before you start and you walk in with a closing statement nobody else has prepared. Three things it is good for, in this order of value. First, the thing you most want them to know that the six questions never asked about. For most people that is either a specific reason for this programme at this firm, or one piece of evidence about them that had nowhere else to go. Second, repairing one answer. If a question caught you cold and you know it, you may briefly give the example you wish you had given. Do that once, keep it short, and do not apologise your way into it, because an apology draws attention to the weakness rather than replacing it. Third, a short genuine close on why UBS and why now. Length is a real decision here. Two minutes of substance is excellent and ninety seconds of substance is better than two minutes padded out, because padding is audible and it is the last thing they hear from you. Three things not to do. Do not read your CV back to them, since they have it. Do not spend thirty seconds thanking them. And do not raise anything negative about the process or your performance.