Quick answer
Find out the format first (coding, system design, take-home or pairing), then spend most of your time on timed practice out loud: core data structures and algorithms, a handful of system-design patterns, and four or five STAR stories for behavioural questions. In the interview, clarify the problem, talk through your approach before coding, and test your own solution.
Technical interviews test two things at once: whether you can solve the problem, and whether people would enjoy solving problems with you. Preparing for both — the knowledge and the way you communicate it — is what turns practice into offers. This plan works for software engineering, data and other technical roles.
In this article
Step 1: find out the format
Ask the recruiter what each stage involves — they usually tell you, and it changes how you should prepare.
| Stage | What it tests | How to prepare |
|---|---|---|
| Live coding | Problem solving, data structures, clean code under time | Timed practice problems, explaining as you go |
| System design | Architecture, trade-offs, scaling | Learn common patterns; practise drawing and talking through designs |
| Take-home task | Real-world code quality, testing, documentation | Treat it like production work; write a short README |
| Pair programming | Collaboration and how you take feedback | Practise with a friend; narrate your thinking |
| Behavioural | Teamwork, ownership, conflict, learning | Prepare STAR stories from real projects |
Step 2: refresh the fundamentals
Most coding interviews draw on a small set of ideas. Be comfortable explaining and using:
- Arrays and strings: two pointers, sliding windows, sorting.
- Hash maps and sets: counting, lookups, de-duplication.
- Stacks and queues: matching brackets, breadth-first search.
- Trees and graphs: depth- and breadth-first traversal, recursion.
- Binary search on sorted data and on answers.
- Big O: time and space complexity of what you write — interviewers almost always ask.
For data roles add SQL (joins, grouping, window functions); for front-end roles, JavaScript fundamentals and the DOM. A one-page reference helps — see our JavaScript cheat sheet.
Try the free Cheat Sheet MakerBuild your own one-page revision sheet of patterns and complexities.Step 3: practise the way you'll be tested
- 1Time yourself. Give each problem 30–45 minutes, as in a real interview.
- 2Talk out loud, even when alone. It feels odd, but it's the skill being assessed.
- 3Write code without autocomplete for some sessions — in a plain editor or on paper.
- 4Review after each problem: what pattern was it, where did you get stuck, what's the optimal complexity?
- 5Do mock interviews with a friend or colleague, swapping roles.
Step 4: a method for the interview itself
- 1Clarify. Restate the problem, ask about input sizes, edge cases and what to return.
- 2Work an example by hand, including an edge case.
- 3Explain a first approach, even a brute-force one, and its complexity.
- 4Improve it if you can, and agree the plan before typing.
- 5Code clearly: good names, small helper functions.
- 6Test it yourself by walking through your example and the edge cases.
If you get stuck, say what you're thinking. Interviewers often give hints, but only when they can see where you are.

System design: a simple structure
For mid-level and senior roles, expect an open question like “design a URL shortener”. A reliable flow:
- Clarify requirements: what it must do, how many users, read- or write-heavy.
- Sketch the high-level components: clients, API, services, database, cache.
- Design the data model and the main API calls.
- Go deeper on one or two parts: scaling, caching, queues, failure handling.
- Discuss trade-offs and what you'd monitor.
Drawing helps you and the interviewer stay on the same page. Practise with simple diagrams.
Try the free Flowchart MakerSketch request flows and architecture diagrams while you practise.Behavioural questions: prepare stories, not answers
Questions like “tell me about a time you disagreed with a teammate” are best answered with the STAR structure: Situation, Task, Action (most of your answer) and Result. Prepare four or five stories that cover a technical challenge, a mistake you learned from, a disagreement, a time you led something and a time you helped someone else.
Try the free Interview PracticePractise answers for a specific job by voice or text and get feedback on each one.Frequently asked questions
How long does it take to prepare for a technical interview?
It depends on how recently you've practised. Many people plan a few weeks of regular, timed practice; if you already code daily, one or two weeks focused on interview-style problems and mock interviews can be enough.
What should I do if I don't know the answer?
Say what you do know, talk through a simpler or brute-force approach, and ask clarifying questions. Interviewers assess how you reason, not just whether you finish.
Which programming language should I use?
The one you're most fluent in, unless the role requires a specific language. Being fast and accurate matters more than the language itself.
What is the STAR method?
A structure for behavioural answers: Situation, Task, Action and Result. Spend most of the answer on what you personally did and finish with a concrete outcome.



