Overly long four hour interview with four sections. Felt like they were trying to make it a Cambridge college interview. Very patronising to anyone who didn’t do undergrad at Cambridge. Got very annoyed at me when I wouldn’t give away secret of a technology I co invented. Didn’t seem like they were fans of anyone who wasn’t exactly like they were, extrapolate from that. If you’re a graduate looking at this, you can do better, trust me, these posh boys have been doing this for years. The technical questions people have listed below are fairly typical, be prepared with board and in depth discussion work if you do go for it.

The process began with a one-hour Teams interview: 30 minutes with HR covering background, motivation, salary expectations and fit, followed by 30 minutes with a technical consultant focusing on a fairly straightforward introduction on both sides. Two days later I was invited to an onsite interview at the Melbourn campus. I drove approximately 150 miles to attend a 2pm interview. Mileage reimbursement was offered at 16p per mile, which disappointingly was well below the HMRC approved rate of 55p. The onsite session included a 30-minute HR interview, a campus tour, and separate technical discussions with two consultants around 2.5 hours in total. The campus is modern, though it may feel remote and isolated for some. Working hours are 9:00–5:30pm, Monday to Friday, with no hybrid flexibility. Annual leave is 25 days plus bank holidays, though three days are reserved for a mandatory Christmas/New Year seasonal shutdown. The technical interviews started with overall look on my CV and then followed by brainstorming on an abstract topic of measuring quantity of an unknown fluid in a unknown container. Some other questions that were discussed: Why does magnetic field change with temperature, how does pulse compression improve time resolution, how would you design a Voronoi partitioning algorithm and how do you collaborate on Github? I am not sure how this style suits them because it prioritizes abstract cleverness and on-the-spot reasoning over proven, deep-rooted expertise. By focusing on first-principles recall rather than a candidate's professional track record, they systematically seems to filter out true domain experts in favor of those who simply excel at a specific cognitive style. Ultimately, this approach favors raw analytical speed and potentially overlooking the specialists best suited for the role. There is a broader tension worth noting: they test for deep first-principles recall across broad domains, yet the role itself involves moving between projects in different fields, which inherently relies on learning on the job.