Explore English
Technical Interviews in English: Explain Your Reasoning
Choose an answer structure, explain technical trade-offs and practise going deeper when asked.
Accessibility options
Match your answer to the question
| If they ask… | Start with… | Then explain… |
|---|---|---|
| “How does an API work?” | A short definition | A concrete request and response, and when it is useful |
| “How would you investigate a slow application?” | The first thing you would check | How you’d narrow down the cause and measure the improvement |
| “Tell me about a difficult bug you fixed.” | Brief context and your responsibility | What you did, what changed and what you learnt |
For a concept, use PREP. Make your point, give the reason, add an example and return to the point. For a practical problem, clarify the goal, explain what you would check first, then describe how you would investigate, test and evaluate. For an experience question, use STARL.
A worked example
Question: “When would you add a database index?”
I’d consider an index if a frequent query is slow because the database is reading many rows to find a few matches. For example, if we often look up orders by customer ID, an index on that field might speed up those lookups. Indexes also take storage and can slow writes, so I’d look at the query plan and compare timings before and after.
Follow-up: “How would you know it helped?”
Answer: “I’d compare the same query before and after the change, using representative data. I’d also check whether frequent writes had slowed down.”
Useful phrases under pressure
- “Could I check one assumption before I start?”
- “My first step would be to find out whether…”
- “I’m considering these two approaches. The trade-off is…”
- “I’d verify the change by comparing…”
- “I haven’t used that directly. Here’s how I’d start investigating it.”
Practise at two levels of detail
Choose a slow dashboard, a failed data import or a feature that behaves differently on mobile. Give a short first answer with the goal, first check and how you’d verify the outcome. Then ask someone to pick one follow-up.
- Why? Why is that your first check?
- How? Which measurement, tool or test would you use?
- What if? What would you do if your initial explanation was wrong?
- Prove it. What would you compare to see if the change helped?
Repeat your answer for a non-technical colleague. Keep the user impact and next step clear.
Continue the lesson
TECHNICAL INTERVIEW | QUICK REFERENCE
Speak clearly. Show your reasoning. Use specific technical examples.
1 Start with the right answer structure
CONCEPTUAL QUESTION PRACTICAL PROBLEM
What is it? → How does it work? → Example → Why or when is it useful?
Clarify → Plan → Investigate / implement → Test → Evaluate
“An API lets applications exchange information through defined interfaces. For example, a frontend requests data from a backend service.”
“I’d reproduce the issue, inspect logs and metrics, isolate the cause, test a fix and check for regressions.”
Match the depth to the question. Start with a clear technical explanation for concepts; give concrete technical steps immediately for practical problems. Go deeper when asked.
2 PREP and STARL: structure your explanation
Use PREP for a conceptual explanation. Use STARL for a real example from your experience. These are guides, not rigid rules.
PREP • Explain a concept STARL • Describe an experience
P — Point: answer directly.
S — Situation: give brief context.
R — Reason: explain how or why.
T — Task: explain your responsibility.
E — Example: make it concrete.
A — Action: describe what you did.
P — Point: conclude with the key takeaway.
R — Result: show what changed.
L — Lesson: explain what you learnt.
PREP example | “Indexes can speed up database lookups. They help the database find rows without scanning everything. For example, an index on a frequently filtered customer ID may improve searches. However, indexes add storage and can slow writes, so I would check the query plan before adding one.”
STARL reminder | For “Tell me about a difficult bug you fixed”, describe YOUR actions and the result, not only what the team did.
3 Quick reminders for each question type
CONCEPTUAL QUESTION → Define it clearly. Explain how it works. Give an example. Mention a relevant trade-off.
PRACTICAL QUESTION → Clarify the requirements. Say what you would check first. Explain your steps. Test and evaluate.
EXPERIENCE QUESTION → Use STARL. Focus on what YOU did and the outcome.
4 Typical questions: what interviewers assess
QUESTION WHAT THEY ARE LOOKING FOR
How does an API work?
Accurate explanation of requests, responses and a practical example.
SQL or NoSQL: which would you choose?
Requirements, differences, trade-offs and a justified choice.
How would you investigate a slow application?
Measurements first, possible bottlenecks, tools and a systematic approach.
What would you do after a production error?
Impact assessment, diagnosis, safe mitigation and verification.
How would you improve this code?
Specific changes, readability, performance and tests.
How would you design a scalable service?
Clarifying questions, architecture, constraints and trade-offs.
How do you know your code is reliable?
Automated tests, edge cases, monitoring and failure handling.
Tell me about a difficult technical problem.
Your contribution, reasoning, evidence of results and lessons learnt.
5 Turn vague claims into technical evidence
TOO VAGUE MORE CONVINCING
DEVACCELERATOR | TECHNICAL INTERVIEW CHEAT SHEET
“I would optimise the database.” “I’d inspect the execution plan, check indexes and look for expensive joins or unnecessary queries.”
“I would make the API secure.” “I’d check authentication, authorisation, input validation, HTTPS and rate limiting.”
“I would test it thoroughly.” “I’d test the happy path, invalid inputs, boundary cases and failure handling.”
6 Example: weak, better and stronger answers
QUESTION “How would you test a login form?”
WEAK “I’d test it to make sure it works.”
BETTER “I’d check successful and unsuccessful logins, input validation and access to protected pages.”
STRONGER “I’d test valid and invalid credentials, empty fields, locked accounts and session handling. I’d also check that errors do not reveal sensitive information and automate key integration tests.”
You do not need to give the longest answer. You need to give an answer that is relevant, specific and appropriate to the question.
7 Useful English phrases
SITUATION WHAT YOU CAN SAY
Clarify “Could I check one assumption before I start?” / “Are there any constraints I should consider?”
Think aloud “My first step would be…” / “I’m considering two approaches…”
Explain a decision “I’d choose this because…” / “The trade-off is…”
Check your answer “I’d verify this by…” / “One edge case I’d test is…”
When unsure “I haven’t used that directly, but I’d approach it by…”
Finish concisely “That would be my initial approach. I can go into more detail if helpful.”
8 Make your reasoning visible
Ask before assuming • Explain what you would do first • Give an example • Mention trade-offs • Test edge cases • Stop when you have answered
9 Common follow-up questions
WHY?
“Why did you choose that approach?” / “What alternatives did you consider?”
HOW?
“What would you check first?” / “Can you walk me through the steps?”
WHAT IF?
“What if it fails?” / “What edge cases or risks matter?”
PROVE IT “How would you test or measure whether it worked?”
YOUR ROLE “What did you personally implement or decide?”
10 Go deeper when asked
IF THEY ASK… YOUR ANSWER SHOULD INCLUDE…
“Why that approach?” A reason linked to requirements, not just personal preference.
“What alternatives did you consider?” At least one option and a meaningful trade-off.
“How would you test it?” Specific test cases, failure cases and how you would verify success.
“What could go wrong?” Risks, edge cases, security or performance considerations.
“What if usage increased significantly?” Potential bottlenecks and how you would investigate scaling.
“Can you be more specific?” Named tools, steps, techniques or a concrete example.
DEVACCELERATOR | TECHNICAL INTERVIEW CHEAT SHEET
11 Technical topics and practice prompts
APIs / backend Explain a REST endpoint; authentication versus authorisation; investigating a slow API.
Databases Explain indexes, joins and how to investigate a slow query.
Frontend Explain state management, browser performance and accessibility.
Testing / debugging Explain unit versus integration tests; reproduce and isolate a bug.
Git / delivery Explain pull requests, code review, merge conflicts and deployment rollback.
Architecture Explain scalability, caching, reliability and trade-offs.
Don’t just say what you would do. Explain HOW, WHY and HOW YOU WOULD KNOW IT WORKED.
Communication guide for English practice. Technical accuracy depends on the role, seniority and specific question.
DEVACCELERATOR | TECHNICAL INTERVIEW CHEAT SHEET
Related resources
Book a session
Practise speaking in a real conversation or prepare for an interview with Katie.
Book a Session →