Final update from StopStalk
This is a slightly emotional one. Bear with me and would love to see you read the complete article to learn from my journey of building StopStalk.
You would have realized from the title that we are sunsetting StopStalk. StopStalk will not be available for usage post 1st August 2025.
Back story of "Why" I built it?
It feels surreal that it has already been 10 years since this. It was back in 2015 when I was in my 3rd year of college in IIIT Hyderabad. Students always used to prepare for internships in the summer of 3rd year. Some of the students also used to prepare for Google Summer of Code. It used to be a common divide that some students were "developers" while others were "coders". I was also one of them while I wasn't best at either in the beginning. I was contributing to an open-source project called Sahana in the hope of applying for a GSoC proposal that summer along with practicing some questions here and there on websites like CodeChef, Codeforces, etc. Leetcode hadn't gotten as popular back then for interview prep (thankfully). I remember it having only a total of 200-300 questions.
Sahana was built on top of a web framework called Web2py. Their developers also developed another framework on top of Web2py with some more abstractions and modularizations called S3. As a newbie, I was struggling to understand which of the features were part of that core Web2py and which of them were implemented in S3. I decided to create my own Web2py application from scratch so that I could contribute more effectively.
It was around that time when I also was stalking my friends' coding profiles by remembering their handles/usernames across every platform. This used to be my motivation for practicing more in the vacations when you're at home and you're also looking forward to getting an internship, which almost always asked DSA questions in their coding tests. I thought about writing a Python script that just crawls the recent submissions of some of my friends and shows it in the terminal. Soon I realized that putting a UI on top of this would be phenomenal (I wish I had that ugly screenshot of this v0 dashboard). I quickly started building a Web2py dashboard on top of this to also help me learn for my Sahana contributions.
Very soon, I realized that I can do a lot more on this database of submissions - give analytics, generate streaks, give GitHub calendar kind of graphs, View submissions, and more. This just made it more interesting. I bought the domain stopstalk.com and started figuring out all the AWS stuff. When you're building something on your "own", you have a high motivation for figuring out everything. I started giving it to my friends, posting on these competitive programming websites, and more that you can send "Friend Requests" on the platform to follow their coding submissions. I just couldn't stop thinking what else I could build on top of this to make it sweeter by the day.
My CGPA continuously started degrading as I got addicted to building this. Fast forward 10 years, here we are discussing my journey and how we scaled from a simple Python script to 130k+ users.
First ever commit of StopStalk
How is the traction like today?
This is something that makes me immensely happy - to build something that people want. For almost the last 5 years, I have done zero social media posts, haven't recommended StopStalk to anyone, haven't actively engaged in any marketing activity whatsoever (we never spent a penny in acquiring any user).
Here is a brief snapshot -
- Total of 130k+ users across the globe
- 8 Competitive programming websites integrated
- 54M+ submissions across all these users
- Multiple colleges using StopStalk rating as their placement criteria
- 90k+ problems fetched from these websites
- A lot of candidates still linking their StopStalk profile in their resumes
How did it help me, personally?
StopStalk has helped me grow personally and professionally as a developer (frontend, backend, devops) as well as an entrepreneur.
Back in the days, I was a newbie developer who had done some self-projects here and there in PHP, MySQL, jQuery.
As part of building StopStalk, I learnt Web2py framework on the fly while contributing to Sahana as well as StopStalk. Figured out how AWS EC2 works, figured out all the DNS stuff, bought a domain, stitched everything together, and haven't stopped learning ever since.
Learnt extreme ownership. I was resolving "support tickets" (emails at that time) within a few hours while I was studying computer networks and humanities and science and what not.
Learnt debugging! This is the hardest skill in tech. Requires patience and sense of urgency at the same time. I still think I am great at debugging because of some of the pressure situations at StopStalk.
All of this experience was extremely overwhelming and fulfilling at the same time. I remember giving the final hiring manager round for an internship interview with BrowserStack - He asked me how did you deploy StopStalk? Explain how you connected your DNS provider? And more in-depth questions around StopStalk only. I guess he got impressed with these specifically since I wouldn't be able to answer these questions if I had just done course projects.
Why did we never monetize StopStalk?
The short answer is we did try. We tried reaching out to companies who could find vetted candidates from StopStalk. We tried reaching out to edtech companies. We tried talking to colleges for premium offerings of leaderboards/score/etc. We tried Ads. And, we also tried donations 🙂
Some learnings around monetization (in no particular order) -
- We were too early for the market
- The contextual data that we used to provide to companies was pretty niche
- College Students did not have money to pay (at least in those days)
- Ads & Donations was the last resort just to pay the StopStalk bills but we stopped them almost immediately once we saw the response.
- There was a skepticism back in the mind that the data that we were providing was primarily built on crawling (we did have official APIs as well in some cases). A platform can any day choose to block us and we will have to rely on skewed data for new users which would have been a mess. In hindsight, this was definitely a right decision.
There could have been some angle for sure in today's market to make some money out of StopStalk but point number 5 was the biggest reason why we didn't aggressively pursue it.
So for 10 years we have been live, with negligible money coming in from an experiment of donations/ads and AWS bills being completely paid from my own savings.
Why are we sunsetting down StopStalk?
As much as I would have personally loved to continue StopStalk, it really doesn't seem to be helping the users in the current state. Here are some of the concrete reasons why we are shutting it down (from highest to lowest priority) -
- The "competitive programming" how it used to be back then has died down in favour of platforms like Leetcode. I personally feel it's unfortunate as now the candidates are just doing "interview prep" and at times just remembering solutions to clear interviews of the top tech companies. 10 years ago, competitive programming was sort of a sport (at least in my college). Of course the end goal was still a job, but students used to really dive deeper and the contest ranks were seen as a thing to improve upon.
- The platforms that we support are also sort of in maintenance mode. This is loosely said but there is a very niche audience who is doing competitive programming for the ICPC sense of it.
- The APIs/crawling scripts of these websites are breaking more often than what I would expect. This has led to a lot of failure, incorrect data and skewed information.
- Because of point 3, my own trust of StopStalk rating has dropped as well. I would not want candidates to be judged for their campus placements on StopStalk rating, not because it is not a good number, but because the rating itself is skewed because the crawling might not have happened for some candidates and might give unfair ranking.
- Monetarily it is not feasible to maintain it anymore. Not just AWS bills, but also the dev effort to work on resolving this. Note that we have also not migrated to a newer framework so we are sort of stuck with a very old web framework called Web2Py (loved it back then though!).
Final remarks
I genuinely love each and every single user who signed up on StopStalk and want to thank every one of you for helping us come this far. It was the fuel to my motivation of continuing to build more features to give a more detailed analysis of a candidate's competitive programming journey.
I am sorry that we have to unfortunately shut it down, but trust me, I would have continued it if the platform was able to deliver the same value as it used to 5 years ago in a student's competitive programming journey.
I also need to thank a lot of my friends and family members who supported me along the way and were also equally proud of me building StopStalk.
Feel free to reach out to me on LinkedIn or over email - [email protected] or [email protected]
