LeetCode has induced instant gratification and made building boring ⚡ Why most engineers today are not thrilled by building something ground up, like their databases, programming languages, game engines, etc? Leaving aside people who are not interested (at all), I see people starting the projects, but giving up pretty soon as they hit their first roadblock. Remember, we can build anything if we persevere enough, but why are we not doing it then? Building something takes consistent time and effort over a long period. But because we are getting wired with quick rewards, there is no fun in struggling and building. Don't get me wrong, doing LeetCode-like questions is fun and has its perks, like, you do become great at problem-solving, writing efficient code, and using the right data structure and algorithms to solve a problem, but the quick feedback loop is kind of messing up with us. Spending hours building something ground up is fun. Spending hours looking at a massive codebase to understand what that one piece of function is doing is fun. Spending hours building that one thing you always wanted to do is fun. Once you are done with enough LeetCode, look beyond those 80/100 lines of code snippets and experience the joy of building something from the ground up. The struggle is real but worth it. ps: I have built my database (dice - redis re-implementation in Go) and a programming language (Revine - visual language for kids). It took me 3-4 months each to build them, but the learnings I got were immense and something that made me a better engineer. ⚡ I keep writing and sharing my practical experience and learnings every day, so if you resonate then follow along. I keep it no fluff. youtube.com/c/ArpitBhayani #AsliEngineering
Elite Gatekeeping to tech from Arpit as always. I used to be the same like you once learning Modern C++. I wrote small compilers, contributed to OSS. Not a single thing helped me get a job. You know what did? Leetcode. I despise leetcode as much as the next guy but today, I am seeing hundreds of talented engineers who are jobless. You think instead of maximising their chances when they get their next interview, they would go built distributed databases? As much as I love your ethos, this is a very first world perspective. I think entrepreneurship is the need of the hour.
Arpit Bhayani I’m trying to understand what you are saying here (I really couldn’t ). Leetcode is meant for improving my problem solving skills. Not meant for giving me the ‘software building’ experience. And doing projects is to improve our Software Development skills. Why are you comparing Bananas with Monkeys here?
Nope. Leetcode and join a proper tech company. Company projects have much better depth, growth and lot of things to learn from than personal projects which no one uses.
Built bitcask in Go couple months ago. Nice blog u had on it.
I come from a Competitive programming background which has slight overlap with leetcode. I want to provide another perspective to this. In Competitive programming we have adhoc problems. Problems that do not rely on existing ideas or use something very unique specifically tailored for that problem. When, I made my initial transition into software engineering, I found my work and the 2 - 3 personal projects I worked on lacking in such adhoc ideas and also at work I had to stare at huge codebases feeling completely lost. I found these to be boring as a result. I felt anything we use to accomplish in software engineering must require some existing ideas, domain knowledge or practices. After a year as an engineer and having worked on another project (an IRC Client) my perspective has changed slightly. Now, I think of software engineering as an amalgation of several small problems / tests that I must solve properly. This has made it more enjoyable but I still prefer CP.
It's because once you have real job experience, most companies won't look at your personal projects. In interviews, interviewers focus on job experience, because it holds more value than personal projects. Most professional software engineers working in big tech companies do their regular work and are not active on GitHub and are doing well.
I see so many databases are popping up every week or so. The problem is not that engineers are not buiding databases or compilers etc. but when you use social networks like linkedin hogged by influencers glamorising leetcode, it makes everyone believe that everyone is just doing leetcode. But engineers who are building those "stuffs" may never be active on linkedin or any other social media. They are just busy building.
I agree. However in my opinion the problem is not with engineers not being excited about building things ground up but the way in which companies hire. One needs a very sound understanding of leetcode problems in order to clear the interviews. Besides that I don't really think that engineers wouldn't be excited if a company asks them to build a simple key value store, give them two-three days and do a code review after the deadline as a part of the interview process. Pretty sure engineers will be very creative and excited about it. This however takes a lot of time that I believe companies don't have and instead go for leetcode. Again not generalizing anything, just putting my view.
I'm not sure of the angle. Are you suggesting that there are now more people in tech that expect instant gratification and thats due to leetcode? I would suggest that is generational and a trait of a person leaving little to do with leetcode. There has to be compassion and a place for each type of person. I, like yourself, come from a generation where tech was in its infancy and had to be created so it became embedded in our interests and stuck with us as who we are. I agree, a very rare quality nowadays.
DiceDB: https://github.com/dicedb/dice Revine: revine.arpitbhayani.me System Design for Beginners: https://arpitbhayani.me/sys-design System Design for Experienced: https://arpitbhayani.me/masterclass Redis Internals: https://arpitbhayani.me/redis-internals My knowledge base: https://arpitbhayani.me/knowledge-base