I applied online. The process took one week.

I will cut to the chase. As recently they've changed the interview process a little bit, and now, you are required to do a take-home assignment. HR and the hiring manager both mentioned it will take you around five hours, but upon reading the take-home requirement, I am pretty sure it's not. It will take only longer than that, like a lot longer. The question is basically to ask you to implement an "on-server" log lookup system. Imagine something like Datadog, Splunk, or Kibana. Basically, you need to build an endpoint to read a file that is over 1GB, and return the lines that contain certain keywords. Parameters include the file path, keywords, as well as the number of rows.

There are some apparent challenges here, like how to dynamically access big content, such as not to load them all into the memory, and also keep in mind the file could grow from an asynchronous process. Also, you will need to lookup multiple files possibly. One requirement is your application needs to be performant, but that's never clearly defined. So, proper solutioning to make it performant will require some serious design and testing. If you are reading the file line by line without any carefully thought-out algorithms, it just won't work. Aside from what's been asked above, they also want you to build a UI (if time permits) on top of this, as well as provide a summary of design on what if it is a primary-secondary cluster setup.

While I was working on the assignment, I tried to ask a few questions like how big of a solution you want this to be built, since it’s a five-hour assignment. Even including a DB sounds heavy they mentioned, so I figured then the solution should be very library independent and overall slim. I pretty much finished all of what they asked for, and my search on a 1GB file which contains 15 million rows will only take three seconds for the worst case, and for some cases, just under 10ms. I wrote up the README.md file to include all the steps on how to set up and run the code, and submitted it to HR. I even included a test collection from Postman to facilitate their testing. I submitted my work in two days, and mentioned that please let me know if anything is unclear, and I am more than glad to clarify.

After a few days, there was no feedback. And today, I sent them an email asking about the progress, and got an email saying they are moving to another candidate. That's it, the end of the email. I am really shocked, like a decision has been made already? Then more questions surfaced, like did they run my code? Did they test it? Or worse, did they understand the code I wrote? Or the algorithm that applied to this solution? It's totally unclear what happened on their end. But either way, it's not a great experience, and I do not appreciate how this ended. It wasted my twenty hours of effort with no effective communication from whoever is reviewing my code. I would never be remotely as angry if my solution didn't work, was flawed, was underperforming, or was just badly written. It's well structured, with separation of classes, and the most important part, it works. Yet you will be rejected with no feedback, even if you ask them what the feedback on the assignment is. Applicants, be aware.

Implement an on-server log lookup system.