Use case · Job interviews

Two weeks to prepare — and on the day, the answer is already in your own words.

You know your own record. What you don't have is instant recall of the project from three years ago, the company's last funding round, and the one figure the interviewer will ask for. Thalassa prepares you from your own CV, the job ad and the company's public record — and during the interview it writes the answer in the first person, ready to say out loud.

The assistant window during an interview — two questions, answered from the candidate's own CV.

Upload the CV, the portfolio, the reference letters, the project notes you never turned into anything. Practice mode then interviews you on them — not on a generic question list. It asks what a real interviewer would ask after reading that CV, including the gap you were hoping nobody would notice: the eleven months between two roles, the title that sounds more senior than the work was, the project you list first but describe least. You answer out loud, it follows up, and the debrief afterwards names the three answers that were vague. Most people discover in that first round that they have never actually said their own strongest example in one piece — they have only thought it.

Every requirement in the posting becomes a question. Where your record answers it, you get the evidence to name: the project, the number, the outcome. Where it does not, you find out two weeks early instead of in the room — and two weeks is enough to build an honest answer for a gap, which is a different thing from being caught by it. The posting also tells you what they are worried about: a requirement listed three times in different words is the reason the role is open. That one gets rehearsed first.

What the sector is arguing about right now, what this company shipped last quarter, who its competitors are and where it is losing to them. Not so you can recite it — reciting reads as preparation, and preparation is not the same as interest. It is so that when the conversation drifts to their roadmap or to the competitor everyone in the room is thinking about, you are in the conversation rather than nodding through it. The briefing takes twenty minutes to read and gives you three questions worth asking back, which is usually the part of the interview people remember.

A 14-day window, no subscription.

Buy the Executive Interview Intensive

During the interview

Written to be said, not to be read.

The answer never arrives as “you could mention that you have five years of experience.” It arrives as “I have five years in marketing, three of them leading a team.” That is a deliberate design decision, hard-coded in every mode: what appears in your window is a sentence you can say, not advice about a sentence.

You know your subject. What costs you the room is the third clause of a long sentence. The answer arrives in clean, natural English — so the interview turns on what you know, not on how quickly you can build a sentence in a language you learned at twenty-five.

There is a dedicated mode for coding interviews: it explains the approach the way a candidate would explain it — the reasoning first, then the implementation, in the first person. You still write the code. What you get is the framing that makes a correct solution sound like a considered one.

It will not invent experience you do not have. Every answer is grounded in what you uploaded, and where your record is silent it says so rather than filling the gap. An interview you win on a fabricated project is an interview you lose in month two. And where a selection process forbids aids, using it is not permitted — that is in the terms, not just here.

Common questions

Job interviews — answered.

Do I need a subscription for one interview process?

No. There is a one-off option built for exactly this: a two-week window that covers the preparation, the practice rounds and the interviews themselves, then ends. See pricing.

Will the interviewer notice?

On screen, no: the window stays out of what you share, there is no sound, no notification and no extra participant in the call. In your behaviour, possibly — someone reading an answer aloud sounds like someone reading aloud. That is why you get a sentence in your own voice and not a script, and why practice mode exists.

Does it work for an in-person interview?

Only if the conversation runs through your computer, or you switch the microphone on deliberately. By default Thalassa hears only what comes out of your speakers — a video call, or a phone call routed through the machine. A phone held to your ear is not heard.

What happens to my CV?

Your knowledge base is searched on your own device. The documents are not uploaded to a server for storage, and the transcript of a conversation is deleted afterwards. How your data is handled.

Two weeks is enough, if you use them.

Install it, upload the CV and the job ad, and run the first practice round tonight.

Start free trial See the one-off option