Arriving with the wrong measurement
I arrived at Imperial College London to study computer science with a familiar kind of confidence. I had done well enough to enter a demanding university and had started to confuse that result with a permanent fact about myself. I thought being clever was something you proved once and then carried with you.
Imperial dismantled that idea quickly. The pace was high, the people around me were capable and the material did not care about my earlier identity. In a recent podcast, I described the experience more bluntly as flunking Imperial. The important part was not the label. It was being forced to confront the gap between potential and performance.
Marks, rankings, admissions and degrees are useful measurements. I had mistaken one of them for proof that the work would keep coming easily. It did not.
I could not bluff my way through confusion
At school, I could often move quickly enough to avoid examining how I learned. At Imperial, speed stopped being enough. Some ideas needed time, repetition and the humility to admit that I did not understand the first explanation—or the second.
When I did not understand something, I could avoid it or sit with it long enough to find the part I had missed. The second option was slower and less comfortable, but it worked.
I became less interested in appearing to know. I asked another question, built a smaller example, read a different explanation or tried to teach the idea back. I still use those methods in engineering work.
Useful learning crossed the classroom boundary
Some of my most practical education happened while building websites and finding ways to pay my rent. The work created a different feedback loop from an assignment. A page either loaded or it did not. A client either understood the result or asked for something else. A promise had a deadline and another person attached to it.
That does not make formal study irrelevant. Theory gives names and structures to problems that experience alone can leave vague. But work made the knowledge answerable to reality. It also made gaps impossible to outsource to a future semester. If I needed to understand something to finish the job, the curriculum was whatever stood between me and the outcome.
I needed both. University showed me problems I would not have found alone. Paid work made me finish what I had started.
Imperial could not do the work for me
Imperial provided capable people, difficult material, laboratories and an environment where ambitious work felt normal. None of that transferred understanding to me by proximity. I still had to do the learning.
That did not end at graduation. Languages, systems and interfaces changed, then the scale and consequences of the work changed too. The degree got me started. It could not keep me current.
I still end up as the person in the room who does not understand something. Imperial made that position familiar, even if it never made it comfortable.
What I would tell a student now
Do not waste a good education protecting the reputation that helped you enter it. Choose some work that may expose you. Ask the question that feels too elementary. Find people who care more about the problem than about performing intelligence. Their habits will teach you as much as their answers.
Failure is not automatically profound. Sometimes it comes from poor preparation or avoidance. Calling it a lesson too early can protect the mistake instead of exposing it.
I flunked at Imperial. I also learned that being clever was not a substitute for practice. The second fact has been more useful than the first.
Public record
These sources confirm the public facts. The rest is my recollection, written in 2026.