How I Fried My Only Laptop for Seeking a Job Offer
Some people say they work best under pressure and I would like to classify myself under that category. If given a challenge that I’m passionate about, I’ll go and scan the lengths and breadths of the task and make sure it is completed before time. The Director of AI of a company reached out to me when I applied for a Deep Learning Engineering role at their company.
The first task that he gave me was video classification into five classes. We were supposed to have 1000 videos in each class. After a lot of errors and obstacles I was able to partially submit the assignment before time. My system, an MSI Raider with RTX 2070 gave me a lot of pain and I ended up wiping up my whole OS thus deleting the 5000 videos I had downloaded that were vital for the project.
I was not satisfied with myself. I had to do something. I reinstalled ubuntu 22.04 along with all the dependencies patiently. Spent sleepless nights until my computer was back to being stable again. It was as if my laptop was going under observation in the intensive care unit and I was constantly nursing it till it regained its strengths.
The very next day, I got an email from them inviting me to attend a video interview. I had to be prepared this time. I had approximately one and a half day’s time. I redid the project and this time the project worked successfully. This made me confident and the interview went pretty well in my opinion.
While interviewer X was interested in the code walkthrough of the new version of the project, interviewer Y was more interested in my past projects in the relevant field. X asked me a few technical questions and I guess he wanted to give me another chance at a project but this time in a shorter timeline.
It was an object detection proposition using YOLOv8 on damaged car parts prediction. YOLOv8 had just been released a couple of months back and the documentation was also not very clear regarding a custom dataset in my opinion. I also had to collect data, and manually label them. It was a huge task. I also had to deploy the model. All this in three days.
Somehow I was able to label and train the model in one and a half day’s time and surprisingly, the accuracy was pretty good. Thanks to my laptop. I was singing praises and complimenting my laptop. The next day i was also able to deploy my model using FAST API. And I pushed the code to my github repository.
But But But, the frontend was not complete. The next day I turned on my laptop only to discover that I had burnt the power IC due to overheating and now I was left with a useless piece which wouldn’t turn ON. I was hell bent on building the frontend using react.js.
I immediately borrowed my father’s Macbook Air which is a 2017 model with limited memory. I had to go through the process of installing all the dependencies and a grand success it was! I was finally able to complete the project.
I am writing this article at 2 am in the morning, and I have my interview scheduled at 2pm. I don’t know if I’ll get the job or not. But the whole process was fun and I’m left with a fried laptop. I think it was one hell of an adventure.
That’s all for now, let’s see how the interview goes. Wish me luck!
BTW, here is the link to my code repo:
Also, special thanks to Nitin Vasanth for inspiring me to write this post.
