Before I started using Craqly, I had done 12 interviews across two months and gotten zero offers. Not zero final-rounds, zero offers. I’d made it to final rounds twice and gotten the “we went with another candidate” email both times within the same week. I want to be clear about that starting point because the results I’m going to describe don’t mean much without it.
Week 1: Mostly awkward
The first thing I noticed was the latency. Responses appear in about 1.5-2 seconds, which sounds fast, but during a live interview that 2 seconds is the exact window where the natural pause ends and a longer pause starts. I was checking the overlay when I should have been making eye contact. I fumbled two answers in the first phone screen not because the suggestions were bad but because I was trying to read while talking.
By the end of week 1, I’d done 2 interviews. One phone screen, one first-round behavioral. The phone screen, I passed. The behavioral, I passed. That’s good data in some interpretations and terrible sample size in any reasonable one.
The thing that did work from day 1: I stopped blanking. I’d had a problem in previous loops where a question like “tell me about a time you disagreed with a manager” would hit me and my working memory would just empty out. The overlay gave me a STAR scaffold visible on screen that I could glance at and return to. Even when I didn’t use the specific wording, having the structure present stopped the blank-out loop.
Weeks 2 and 3: Where it actually started to help
I passed all 4 behavioral rounds I took in this window. I also did my first technical round with the tool running, and that was more complicated.
The coding suggestions were directional, not complete. When the problem was something I’d seen pattern-similar problems before, the suggestion would point me toward the right approach and I’d fill in the actual code. When the problem was genuinely unfamiliar, the suggestion was sometimes wrong in ways that would have been obvious if I’d thought about it, and not so obvious when I was under time pressure. I used one suggestion that led me down a path I had to abandon about 8 minutes in. The interviewer waited it out. I didn’t get that offer.
Behavioral rounds are where the tool adds the most reliable value, at least for me. System design rounds are second. Coding is third, and that’s probably where it should be. You need to understand what the tool is giving you, and coding is the domain where misunderstanding the output has the fastest and most visible consequences.
What didn’t work
I had one case study round, and I did not pass it. The format was a 40-minute business case with a written prompt and a presenter. Craqly’s suggestions were generic enough to be useless. The tool is built around structured question-answer formats; open-ended analysis with no defined question type doesn’t fit that pattern well.
Post-interview logs, where the app shows you what suggestions appeared and when, are genuinely useful for identifying patterns in your weak spots. I found out I was consistently underperforming on “tell me about a failure” questions because the STAR framework breaks down a bit when the story has to end with the failure still being partially unresolved. I knew that intellectually. Seeing it in a pattern across 4 interviews made it concrete in a way I hadn’t expected.
The results at the end of 30 days
8 interview processes. 2 phone screens (both passed). 4 behavioral rounds (all passed). 2 technical coding rounds (1 passed). 1 case study (failed). Final outcome: 3 job offers, which was 3 more than I’d gotten in the previous 2 months.
I don’t know how much of that is the tool and how much is the fact that my second 8 interviews happened after I’d already done 12 and was simply better at interviewing. I think it’s some of both. The Stack Overflow 2024 Developer Survey found that 62% of developers use AI tools in their day-to-day work. Getting comfortable with AI assistance in high-pressure environments is probably a skill that transfers, not just a shortcut.
Would I use it again?
Yes, with the caveat that it changes how you prepare, not how much you prepare. I found I spent more time before interviews thinking about which parts of my experience matched which question archetypes, knowing the overlay would help me retrieve and structure in the moment. Less time drilling rote answers. The tool shifts prep toward understanding and away from memorization. That’s probably good for interview performance and definitely good for the actual job after.
The candidates who get the most out of it are the ones who already know their stuff and need help with the retrieval and structuring under pressure. If you’re genuinely unprepared, the overlay surfaces that gap faster than it covers it. The interviewer asking a follow-up to a suggestion you don’t understand is a worse position than not having the suggestion at all.
After 30 days, my view is that it’s a solid tool for the thing it’s primarily good at. The honest version of that sentence includes the fact that the thing it’s primarily good at is behavioral rounds, not the full loop equally.