I was at university working on a side project: I wanted to have my own movie torrent website. The idea was to list movies and under each movie list the best ava

I was at university working on a side project: I wanted to have my own movie torrent website. The idea was to list movies and under each movie list the best available torrents. This is now a pretty common feature, but at the time trackers only had a list of file going like Titanic.1998.DVDRiP.Proper.MyGroup. I was able to fetch the torrent, but the hard part was linking each torrent the correct movie (by IMDb ID). I tried for ages to get it right with reg exp to extract the movie title from the torrent name, use the IMDb API, and so on until I found the solution. The solution, was to just simply take the torrent name, extract the year as the last 4 digit sequence, remove all regular tokens like dvdrip, TS, xvid and so on and perform an automated google search on the string: IMDb "Everything else" (Year). and extract the ID from the first result. It took me less than an hour to implement, and gave absolutely glorious results, matching movies with typos, rejecting non movie torrent and generally getting a fantastic accuracy. I went on to release the website, which went on to become semi famous in the field and the consequences for me were : - Sometimes complex problems actually have simple solution, think outside the box, you don't have to do everything yourself. - I actually earned money from some ads I put there, which made uni life a lot funnier. I also experienced what it's like to put a ton of efforts in a product only to get a very slow growth. And then see "something" happen, and you get rewarded for all of that. In a way I felt like a founder. - I maintained the website for years. I was able to apply most of what I learned in my lessons on an actual product, learn new stuff on my own, and faced challenges like scaling that my peers simply could not grasp. I finished uni with top grades and a significant reason for that was my illegal torrent website. - I got a ton of recognition for my work from friends and family. Even met someone at a bar that wanted to show me the cool website he found when we started discussing movies. One of my bets memories. I started working at BigCo and did not get that kick anymore, which prompted me to change path and work in startups. Really happy with that choice. The code itself was simple, not really beautiful or efficient, non scalable, completely against any TOS you can dream of, and the goal was to aid movie piracy. It dod not generate million of USD. It probably did not change the life of many, but it changed mine