Giving Back: My Journey to 375 TC

tl;dr old TC of 145k -> new TC of 375k Blind changed my (professional) life. This community is an indispensable resource of unique know-how that is difficult-to-impossible to find on other online venues. In this post, I will do what I can to return the favor and share with you my journey from “technical employee” to “startup SWE” to “faang SWE.” I can’t say how much of what I did is reproducible for you, but I can say that the success stories you see here are real and attainable with dedication to the grind. This is just what happened to work for me; maybe it can help you on your path as well 🚀 It started somewhat accidentally, by mistakenly clicking a random ad on Reddit for TripleByte, sometime in 2018, and then getting sucked into their email funnels. I had a couple YOE doing data analytics work and knew enough Python & SQL to be dangerous. Passing the initial online quiz for TripleByte was easy enough, but did I really have what it takes to make it out in Silicon Valley like they suggested? TripleByte was going to conduct their version of a remote, on-site interview to find out. Like any self-reliant wannabe, I began Googling things like ‘how to pass TripleByte interview.’ I got a little smarter and improved upon the initial query via ‘site:reddit.com TripleByte interview,’ which led me to the datapoint @ https://bit.ly/3KB7RQy. 🤔 Hmmm… ”used car” is a pretty specific term, let’s see what Google has for ‘"used car" "triplebyte"’. “Wtf is Blind,” I asked, clicking the link from the Google search results. Idk, but the post @ https://bit.ly/3pZejHv is way too fucking specific to not be real. Better bookmark this bad boy, I thought to myself. I decided to take a leap of faith and treat this as my source of truth for what to expect. For the next couple of days ahead of my TripleByte interview, I read everything I could about those concepts. And it worked. Somehow, this random post – on this random website I had never heard of – had enumerated nearly every piece of knowledge I was quizzed on in the interview. That’s one domino down, but this just meant that I now had access to TripleByte’s platform to schedule “real interviews” with “real companies.” I “liked” some companies on TripleByte; some companies “liked” me back. Phone screens were scheduled. I crafted a pitch on “who I am, what I’ve done, and where I’m going.” Phone screens were had. The shake out from all those phone screens was five on-sites in the Bay Area in five days. The interviews were rough. One interview sent me home early. It sucked. Most companies didn’t make an offer. But one felt good enough to take a chance on me. It’s better to be lucky than good, and you only need to be lucky once. Suddenly I was moving to San Francisco. The pay was 🥜 relative to the COL, but it was enough to pay rent and be referred to as a “software engineer.” The time I spent at that startup was great. The company gained a solid employee who busted ass everyday to get shit done, and I learned a lot & built valuable skills with great people. Throughout this time, I kept browsing Blind, which led to the eventual schlepping of Leetcode. It’s not uncommon to come across near-hyperbolic success-stories on Blind along the lines of “How I did 200 problems in 30 days and cracked faang.” I can’t claim to have done 200 problems in 30 days; in fact, it took me well over a year to hit those numbers. But what I can say is that I never stopped never stopping – and that’s the key 💪 One series of posts was particularly inspiring & instructional for starting to seriously grind Leetcode, and I’ve compiled these “how I cracked Google L6” posts by user “flung2” @ https://bit.ly/3AZziAo. I knew that I wasn’t cut out for L6 any time soon, but aim high and miss low, right? After I had done 100 problems or so, via a combination of Blind75 & Grokking the Coding Interview, I bought a coaching package from interviewing.io for five sessions with a Google interviewer. The coaching package was worth every penny. While I had become decently adept at “cranking out leetcode,” there was another plane of performance involved in “real” interviews with another person on the other side of the keyboard. The coaching package was invaluable for bringing me up to speed with interviewing soft-skills, e.g. how to deftly get your interviewer to give you clues when you get stuck. From mocks and various sources on the internet, I eventually distilled coding and system design interviews down to a formulaic “flow.” If I had trouble with a problem on Leetcode or struggled during a mock, I often found that the source of the failure could be found in a deviation from the intended flow. These flows are @ https://bit.ly/3TOHD1c. Another perspective that the mock interviews illuminated was that I had not learned some of the patterns as well as thought. More specifically, I had been investing time into thoroughly understanding the pattern behind the problem, but I had not been systematic about reviewing the pattern after the fact. I mentioned this to a buddy who suggested that I look into SRS (Spaced Repetition System) memory management technique 🧠 Anki is a popular tool for managing SRS flashcards, but I found that there was poor support for code highlighting, which was important given that I wanted to commit algo/ds patterns to memory. With a little bit of Googling, I eventually stumbled across a VS Code plugin called Recall @ https://bit.ly/3Rilfve, which allowed me to make and manage flashcards by writing simple markdown. The end result is a library of flashcards & Python templates for coding & system design interviews, which I have made public @ https://bit.ly/3CM8Qvq and @ https://bit.ly/3B1fa0E. At a certain point, I realized that it was time. Time to put down the leetcode practice, sack up, and take my shot at faang. I said fuck it and put my two weeks notice in at work; let’s stack the chips and put them on the table. Yolo 🎲 I had been 90% focused on leetcode preparations, so I hunkered down with “Grokking the System Design Interview” and “Designing Data Intensive Applications;” (the DDIA audiobook was a surprisingly effective medium). My interview coach helped me with a referral to Google; Amazon recruiters on LinkedIn were more than happy to see me reply to their messages for once. I scheduled a handful of other interviews to have competing offers and something to fall back on in the event that I didn’t get any offers from faang. Now, everything was in place… except for Amazon’s Leadership Principles. I got my Nalgene out and started chugging. All. The. Kool-Aid. All. of. It. I read up on the LPs and made a matrix/spreadsheet of my experience/projects versus the LPs, marking with an X to indicate that the project was an example of the relative LP. Then I sorted & filtered projects with the highest LP count, (ensuring that I had at least 3 stories per LP), wrote out the details of each project in STAR format, and finally condensed everything down to a single page of a simple bullet-point list with “Date - Project Title - One-liner Note.” The final, simplified version of LPs was used as a reference during live interviews. Preparing this way for Amazon’s behavioral interviews meant that I could pass other company’s behavioral rounds without breaking a sweat. All of my LP resources were compiled & archived @ https://bit.ly/3wLRPxo. Once my calendar indicated that Company-XYZ was next on the interview docket, then I would work exclusively on their problems. Filter Leetcode problems tagged with Company-XYZ, sort by most frequent, and then do as many as possible, top to bottom. I saw several repeat problems in interviews by following this strategy. After everything was said and done, I was successful in my attempt to crack faang. Unfortunately, I got caught by the Google 🧊 after passing the phone screen, so they’re off the table (for now). But I passed Amazon’s SDE-II interview with flying colors and was able to leverage competing offers and negotiate to a 375k TC with the help of interviewing.io’s negotiation service. The end 💰 AMAA

Practice interviewing with engineers from top companies, anonymously.

interviewing.io