Some notes from my journey in the wilds of finding a new job in tech, in 2025.
My search centered on roles like SRE, DevOps, Cloud Infrastructure, Platform Engineering, and similar roles at Senior and Staff level, 10+ YOE. I did try for a couple Principal levels as a stretch, but, never heard back after applying. Basically, the things that make computers go beep boop reliably, and enable infrastructure consumers easy, reliable, repeatable paths with low/no friction.
This is in no way exhaustive, but, some general thoughts about my own journey. I don't really have a true way of knowing if any of these steps, or even if the collection of all the steps, is "the right way". It's just what I did, which from my perspective worked well. I was able to consistently get screening and hiring manager calls out of this. However, even with all of this, my rate of contact after an application was barely 50%. I hope you find value in this somehow - applying and interviewing is full of high competition right now. My journey, which I pushed myself very hard on, took ~10 weeks to my first offer.
A personal opinion - companies doing what I'll brand as "extreme" interviews (>6 interview steps after the first screening call), I mentally depriorized at least a little. To me, needing that many interviews to make a decision was itself a signal, and not a positive one. There was one instance where a company had 8 interviews after the initial screening. I had other concerns going into that one, and when I heard that process, I politely removed myself from their consideration. This is a priviledged position for me to take, and not everyone has the luxury to do this, which I fully acknowledge.
AI is prevelant everywhere. On LinkedIn, if you scroll three pages, you'll find people telling you to definitely use AI, definitely don't use AI, and everything in between. My personal method was to not use AI for basically anything. A couple of interviews I did involved using AI with prompting (at the interviewer's request/insistence), in particular during live coding type loops. I felt that AI was more distracting, almost in a way that felt like another person was there in the interview, and it never really felt like a positive addition. Later, if an interviewer would offer that I could use AI, at best I would use auto-complete while still very much talking through my solution, trade-offs, and certainly calling out things I didn't like about what the AI did, demonstrating that I didn't just blindly accept it's contribution (even if it worked fine). I would certainly suggest asking if you are unsure, and if they say to not use AI, don't even try cheating.
I'd often spend 15-30 minutes tweaking my resume to align with Job Descriptions (JDs), without overstating my abilities or anything like that. At least 50% of applications got a very tailored resume, but some were "close enough" that I could reuse resumes with little or just tiny alterations. I have a two-page document with things I can just quickly copy/paste for my last 3 positions. People generally didn't seem to care about things more than ~5 years ago — I could probably trim older jobs back even more at this point. I always made sure to keep my resume right around 2 pages, never more.
My resume doesn't have a free form text section (no "about me", "objective statement", "summary"), nor do I have a list of technologies/languages. I have my name, my phone, my email address, then jump straight into work history. Each work history has meaningful data below it - what I did, what the outcome was, the impact from it (with as much supporting numbers as possible - such as $ saved/made, % of alerts reduced), and any tools/stacks I used. Quick and poor example would be like "Tech Lead for delivery of SuperMassive Project, generating $16 trillion in ARR, through pairing with peers across teams, building the Azure cloud-native solution on VMs and k8s with new observability primitives"
I wrote new, individualized CLs for each application, at least for the first bunch of weeks I was looking. I tried to match the company's tone a bit... if the JD and career page was casual/laid back, I'd match that, or if it was stuffy and formal, I'd match that. I tried to keep CLs to about 200 words, and key into things from the JD, too. A lot of places I talked to mentioned my cover letter, either in the first screening call or the call with the HM, so it felt worth it. Towards the end I definitely recycled CLs though, as I got real tired of writing brand new every time.
Other "tips" I stumbled into (sorry, this got long as I was typing, read at your own risk)... I have no means to get data or anything to "prove" these, so, take it all with your favorite amount of salt.
-
Schedule interviews at a time that's good for you. I tried to accommodate, and at times that meant doing early or otherwise poorly timed interviews, which wasn't putting my best foot forward. I found that doing interviews just after lunch time was ideal for me, and seemed to also be the best for the interviewers as well.
-
I'd also suggest not doing that interviews rapidly. It sucks, and burn out set in real fast for me. One or two interviews per day would be healthy and still a brisk pace, if you have the time for that. I did as many as 4 in a day, and that was a really bad idea.
-
A friend shared a spreadsheet to track applications and status, which I've tweaked for my own use, that I use to track things (Template available here which you can copy with File > Make A Copy).
-
I also keep extensive notes on my interviews and the company. Each company has their own page for notes, and I'd spend 5-10 minutes writing everything down I could think of after every interview. This was really good for the "Do you have any questions for us" phases and just helped general organiation as I was moving through processes. "Ok I just did third interview, is this the place with 4, or 10? Lemme look...". It was also nice to just reference to gauge my own interest in an opportunity outside the pressure of an interview. I ended up opting out of several places after reviewing my notes.
-
I did a lot of up front research. I'd often do a quick read for a few minutes (checking for any quick red flags / deal breakers), then apply to try to get in early for new opportunities. Most evenings I'd be spending 2-3 hours reading up extensively on every place I applied to. Got to know their products, checked their incident pages, tried to find people on LinkedIn, read Glassdoor, etc. This paid dividends in the first screening call and hiring manager calls. I got many compliments on how much I knew about companies going into interviews. I also just found some new tools and tech stacks to actually use and learn along the way!
-
Be nice to recruiters. You may end up ghosted, or otherwise not treated in the most respectful way. That sucks, and I can sympathize with those experiences. But! I treat all of them as well as I'm able to. Thank them for their time, and be good at communicating with them. They're often a bit of an internal advocate for you, so the easier you make their job and the more likable you are, it seems the more they'll go to bat for you. Send meaningful/insightful, but brief, follow ups after each "round" you go through — I try to get these sent something like 8-12 "business hours" after the interview.
-
Startups and smol companies seem the easiest to get attention with. You can often find the hiring manager on LinkedIn, and a lot will respond to DMs there, especially if they have the blue "HIRING" banner. There's big companies that I applied to in May and June which I still haven't heard back, and still show my application as "active" on their candidate portal. I was also mostly focusing on small companies for my next move (I wanted to intentionally move to a much smaller company), so, this may just be selection bias at play.
-
Use the "Last 24 hour" search in LinkedIn. Getting in early seems to be one of the most meaningful factors to getting a response. A lot of the companies that never replied to me had their listing up for several days or even weeks. If I got my app in the first day, my hit rate was way higher.
-
Have STAR responses written out and rehearsed, but do not go on auto-pilot when prompted. I have 11 total written out and that I could just speak through without even thinking. I'd suggest at least try to have 3 or 4. Tying these responses into the company you're interviewing at, or something else in context of the current discusion certainly helps make it real, shows you were listening, and that you can directly tie your experience into their needs. You definitely don't want to go on a 5 minute monologue. There's always going to be follow up questions and people digging in deeper, so don't have fictional answers here.
-
I was spending about 2 hours nightly working on my non-AI coding skills. Those have atrophied big time, and live coding interviews really put a spotlight on that. As much as I hate leetcode, it was beneficial for easy/medium puzzles to jump into quickly. Exercism is also fun. Learn the WHY as well as the implementation as you go. If you have a pet project, work on it more. Currently, pretty much all live coding interviews do not allow AI tools -- I expect this to change over time, so if you've stumbled on this in 2026, it may already be out of date.
-
In addition, I also read through books again like Cracking The Coding Interview, Head First Design Patterns, Groking The System Design Interview, and so on. While some of these are dated (though there are newer revisions than the copies I have), many of the concepts and foundational aspects still very much apply. There's also a lot of other useful, free, online resources, and maybe I'll update this with a resources section later when I have a chance.
-
System Design interviews are everywhere. I think only two places I interviewed at didn't have a sys design stage. Practice this over and over again, even if it's a core part of your job/experience. Don't just slam through a solution. There's lots of good references on how to do well at this interview.
-
Searching by title is wild and difficult. I still haven't cracked how to really SEARCH for a job, so a lot was scrolling the "Last 24 hour" list and opening a gazillion tabs. Might be a "me" problem, but I found fits in SRE, DevOps, "Backend", "Platform Engineering", "Cloud Engineering", and other titles. Also found a lot of really big misses in the same titles, so building an eye for finding the relevant parts of the JD helps for filtering.
-
For my roles, there were a lot of live debug/troubleshoot type interviews. So practicing fundamentals in Linux, k8s, networking, etc. were good to brush up on. I do this almost naturally anyway as I maintain a fairly large home lab, but I still spent focused time on learning. One example was a place really wanted me to come up with a bash one-liner to "find the count of each word in a file". i.
grep -oE '\w+' file.txt | tr '[:upper:]' '[:lower:]' | sort | uniq -c | sort -nrwas my generalized solution, and thankfully having my lab and experience, I was able to do that nearly instantly which they appreciated, while explaining a key trade-off (can you spot it?) -
Drink water. Before an interview, spend at least 5-15 minutes doing nothing, take deep breaths, and try to go in as even and calm as you can. It's going to be stressful and uncomfortable, so try to go into it clear and focused. Don't run and grab a drink 1 minute before the Zoom starts - you might trip over a cat going down the stairs. Purely hypothetical situation that I definitely didn't have happen to me.
-
For take-home things, go above and beyond. Write some tests and say why you did with comments. Put a comment explaining a tradeoff (ex: why did you choose fixed window over token buket?) Write a README. Show your thinking and intent behind your execution. Even these take-home exercises are a conversation. Your solution is of course important, but just like High School, "show your work" helps have that async conversation with whomever is evaluating what you submit.
-
"Give us a presentation on a project you lead" seems kinda common now? This was a brand new flavor of interview for me. My opinion right now is that I like it a good bit more than Live Coding, but, it's a new kind of stress. I had to do it something like 8 times. I was able to reuse a presentation a couple of times, but, not frequently. It was also a little challenging to keep in mind that there may be areas I can't go deep on for security, IP, or other reasons, but most interviewers were very understanding of that. These were often an hour, and I aimed for my presentation to take 25-30 minutes, allowing for them to question and go deeper, and of course my own questions afterward.
-
Lastly, be ready for "the usual" questions. I was surprised how many places still ask old standard type questions — going into my first couple of interviews, they caught me off guard and it definitely showed.