Home / Blog / Are you actually interview-ready? A mock-interview rubric

Are you actually interview-ready? A mock-interview rubric

PrepHike · 6 min read

Most engineers decide they are ready when they feel calm and have revised a few topics. That is a guess, not a gate. A real mock interview tells you something a feeling cannot: whether you convert under pressure. Here is the rubric we use to score readiness, and how to run it on yourself.

A mock is a gate, not a formality

The question "am I ready for interview" usually gets answered by mood. You revised arrays, you reread your own resume, you feel less nervous than last week, so you start applying. Then a real interviewer asks a follow-up and the answer wobbles, and you find out in the worst possible place that you were not ready.

A mock interview exists to move that discovery earlier, where it is cheap. Treat it as a gate. You do not apply to your target companies until you clear it. This matters more in India than people admit, because the cost of a failed real interview is not just one round. Strong companies often put you on a six to twelve month cooldown before you can reapply, and your first offer sets the anchor for your next package. Walking in underprepared does not just lose one shot. It can lock you out of a 30 LPA door for a year and keep you anchored to your current pay.

So the bar for a mock is simple and strict: it has to be live, it has to be scored against something objective, and you do not get to pass yourself on vibes. An AI mock is fine for extra repetitions during the week. It is not the gate. The gate is a real human asking real follow-ups, because live pressure is exactly what you are training for.

The readiness rubric: six dimensions, scored 1 to 5

We score six dimensions. Four are technical, two are behavioral. Each one runs from 1 (fail) to 5 (ready), with 3 as borderline. The pass rule: your average is at least 4, and no single dimension sits below 3. One score of 2 fails the gate even if everything else is a 5, because interviewers reject on the weakest signal, not the average.

1. Problem scoping

A 1 jumps straight to code with no questions. A 3 asks for some clarification when prompted. A 5 scopes the inputs, constraints, and edge cases before writing anything: "Can the array be empty? Are values within int range? Is it sorted? What is the expected size, hundreds or millions?" Scoping is the cheapest signal of seniority there is, and most underpaid engineers skip it because at work the requirements are already known. In an interview, the person who scopes first reads as the person who has owned design decisions.

2. Communication

A 1 is silent or rambling. A 3 explains after being asked. A 5 thinks aloud in a structure the interviewer can follow: "My first thought is brute force at O(n squared). I think we can do better with a hash map. Let me check that before I commit." This is the dimension that quietly decides offers. Two engineers can write the same correct code, and the one who narrated their thinking gets the higher band, because the interviewer could see the reasoning instead of guessing at it.

3. Correctness

A 1 is wrong or stuck. A 3 gets there with hints. A 5 is correct and handles edge cases without being reminded: the empty input, the single element, the duplicate, the overflow. Note that correctness is one dimension out of six, not the whole game. Plenty of engineers obsess over solving the problem and ignore the five other things being measured at the same time.

4. Complexity and trade-offs

A 1 never mentions complexity. A 3 states Big-O when asked. A 5 reasons about trade-offs unprompted: "This is O(n) time but O(n) space. If memory is tight I can do it in place at the cost of mutating the input. Which matters more here?" For system design the same dimension shows up as capacity and failure: a 5 estimates load, picks components for reasons, and handles what happens when a node dies. Our system design checklist breaks that down further if design is your weak track.

5. STAR structure and specificity

This is the behavioral half, and it is where strong engineers lose money fastest. A 1 is a vague story with no arc and no personal role. A 3 has loose structure and some detail. A 5 has a clear situation, task, action, and result, owns a specific personal contribution, and lands on a real number: "Cut p95 from 1.2s to 340ms," not "improved performance." If your project stories come out flat, two posts will help more than this one: the 30-second method for explaining a project and STAR stories that don't sound rehearsed.

6. Composure under follow-up

A 1 crumbles when challenged. A 3 recovers slowly. A 5 holds up and defends a decision calmly, or concedes cleanly when the interviewer has a point. This is the dimension you cannot fake alone at your desk, which is the whole reason the gate has to be live. The interviewer pushes: "Why not use a queue here?" A ready candidate either explains the choice or says "Fair point, a queue would handle the spike better, here is the tradeoff," without the panic that reads as not knowing your own work.

How to self-assess before you book a real round

You cannot fully score yourself, because composure and communication need an outside observer. But you can get most of the way with an honest dry run. Record yourself solving one medium problem out loud, on a clock, with no pausing. Then watch it back and score the four technical dimensions cold. Did you scope before coding, or jump in? Could a stranger follow your narration? Did you mention complexity without being asked? Most people are shocked at how much silence and how little structure shows up on the recording.

For the behavioral half, take your three strongest projects and time yourself telling each story. If you cannot land situation, task, action, result, and a number inside two minutes, you are not ready on that dimension, no matter how well you know the work. Write the number down for each story first. If you do not have a metric, that is a gap to fix before the mock, not during it. And before any of this, be honest about whether your target band is even calibrated: the market value audit tells you what "ready" needs to clear for the salary you are aiming at.

Be strict with your own scores. The temptation is to round a 2 up to a 3 because you "basically knew it." Interviewers do not round up. If you keep landing 3s, you are borderline, and borderline loses to the candidate who prepared properly. Send yourself back to drill the specific weak dimension, then re-test. This is the exact discipline of the Fire-test phase of the SHIFT method: you do not move to applying until the rubric clears, because a real failed interview costs you the opportunity and the cooldown.

What a clear gate actually buys you

When you walk into a real interview having already cleared this rubric in a live mock, the round stops being a test and becomes a performance you have rehearsed. You scope on instinct. You narrate without thinking about narrating. You hit your numbers. And when the follow-up comes, you have already been pushed on harder questions in practice, so you hold steady. That composure is what reads as senior, and senior is what gets paid.