Interview scorecard
Full-Stack Engineer
Score each skill from 1 to 5. A 1 is below the bar for this role, a 3 meets it, and a 5 is above it. The examples show what each sounds like. Use 2 and 4 for answers that fall in between.
System design
Breaks a vague product need into frontend, backend, and data parts and weighs trade-offs between them.
Questions
- Walk me through a feature you built end to end. What did you need to know first?
- Tell me about a time your design hit a limit. What did you change?
- Describe a design choice where you picked simplicity over flexibility. Why?
1: Below the bar
Jumps straight to code without asking about requirements. Can't say why they chose one approach over another.
3: Meets the bar
Reaches a workable design across frontend and backend. Names main trade-offs but misses growth limits or failure cases.
5: Above the bar
Decides what belongs in the browser, the server and the database with reasons, and explains what would break under growth.
Frontend craft
Builds interfaces that are usable, accessible, fast, and easy for other engineers to maintain.
Questions
- Tell me about an interface you built that you're proud of. What made it good?
- Tell me about a time a page felt slow or broken for users. What did you do?
- How do you make sure a screen works for keyboard users and different screen sizes?
1: Below the bar
Describes only how the screen looked. Never mentions users, accessibility, speed, or how the code will be maintained.
3: Meets the bar
Names real user needs and one or two practices, like reusable components or checking mobile. Gives few specifics about results.
5: Above the bar
Ties choices to user needs, with examples on accessibility, speed, and maintainability. Describes how they checked it worked and what they measured.
API design
Designs clear, consistent API contracts that other developers can use and that survive change.
Questions
- Tell me about an API you designed. Who used it, and what did they need?
- Tell me about a time you had to change an API others relied on.
- How do you decide what an API should return when something goes wrong?
1: Below the bar
Describes endpoints as whatever the screen needed. Has no plan for errors, naming consistency, or what happens to existing users.
3: Meets the bar
Designs consistent, sensible endpoints and handles common errors. Considers other users' needs but plans for change only when prompted.
5: Above the bar
Starts from the consumer's needs, keeps naming and errors consistent, and explains how they changed the API without breaking existing users.
Debugging
Isolates the cause of a bug methodically and confirms the fix works before moving on.
Questions
- Tell me about a bug that took you a long time to find. How did you narrow it down?
- Describe a bug that only appeared in production. What did you do first?
- Tell me about a fix that didn't work. How did you change your approach?
1: Below the bar
Changes code by guesswork and never reproduces the bug. Can't explain how they knew the fix worked.
3: Meets the bar
Reproduces the bug, narrows it down step by step, and checks the fix. Finds the cause with some trial and error.
5: Above the bar
Forms a clear hypothesis, tests it quickly, and finds the root cause. Adds a test or check so it can't return.
Ownership
Takes responsibility for outcomes beyond assigned tasks, following work through to real results for users and the team.
Questions
- Tell me about a time you saw a project through after your part was done.
- Describe a time something you shipped caused a problem. What did you do?
- Tell me about a time you fixed something nobody asked you to fix.
1: Below the bar
Stops at the assigned task. Blames others or circumstances when things go wrong and names no follow-up action.
3: Meets the bar
Finishes tasks and fixes own mistakes when asked. Follows up on results sometimes but rarely acts beyond the assignment.
5: Above the bar
Acts without being asked, tracks results after launch, owns mistakes openly, and describes lasting fixes that helped users and teammates.
Overall recommendation
Decide on your own, before you compare notes with the rest of the panel.