TECH ENGLISH

Technical Interviews

Ace your tech interview in English — learn how to explain your thought process, ask clarifying questions, and discuss time complexity clearly.

Practice Tech Discussions

Scenario Context

You're in a 45-minute coding interview at a top tech company. The interviewer gives you a problem: design an algorithm to find the shortest path in a weighted graph. You need to clarify the problem, think aloud as you develop your approach, explain your code, and analyze the time complexity.

Why This Matters for Engineers

In a technical interview, how you communicate is as important as whether you solve the problem. Interviewers evaluate your thought process, your ability to ask clarifying questions, and how you handle ambiguity. Many talented engineers fail interviews not because they can't code, but because they can't explain their approach clearly in English. Thinking aloud fluently while solving problems is a skill that requires specific practice.

Essential Phrases

“Before I dive in, let me make sure I understand the problem correctly.”

Clarifying requirements

formal

“Just to confirm — can the edge weights be negative?”

Asking about constraints

neutral

“My initial approach would be to use Dijkstra's algorithm here.”

Proposing a solution

neutral

“Let me think through the edge cases before I start coding.”

Considering edge cases

formal

“I'm going to start with a brute-force approach and then optimize.”

Starting simple

neutral

“The time complexity of this solution is O of n log n.”

Analyzing complexity

formal

“Let me trace through this with a small example to verify.”

Testing the solution

neutral

“I realize there's a bug — let me fix this edge case.”

Catching a mistake

casual

“If I had more time, I'd optimize this by using a min-heap.”

Discussing improvements

neutral

“Could I use a hash map here to get constant-time lookups?”

Exploring data structures

neutral

Technical Pronunciation

Word	❌ Common Error	✅ Correct	Tip
algorithm	AL-go-rith-um	AL-guh-rith-uhm	Four syllables — the second syllable is a quick 'guh'.
Dijkstra	dij-KES-tra	DIKE-struh	Named after Edsger Dijkstra — sounds like 'dike-struh'.
recursion	ree-KUR-zhun	rih-KUR-zhun	First syllable is a quick 'rih', not 'ree'.
paradigm	PARA-dig-um	PAIR-uh-dime	Ends in 'dime', not 'dig-um'. The 'g' is silent.
memoization	mem-or-eye-ZAY-shun	mem-oh-ih-ZAY-shun	Not 'memorization' — it's 'memo-ization', from 'memo'.
tuple	too-PULL	TUH-pull (or TOO-pull)	Both pronunciations are accepted. The first is more common in American English.

Written vs. Spoken English

Engineers often write one way on Slack or GitHub, but speak differently in meetings. Here's how to translate.

Explaining complexity

Written (Slack/PR)

Time complexity: O(V + E) log V using a binary heap implementation.

Spoken (Meeting)

“The time complexity is O of V plus E, times log V — that's because each heap operation takes log V time.”

Describing approach

Written (Slack/PR)

Implement Dijkstra's algorithm with a priority queue to efficiently extract minimum distance nodes.

Spoken (Meeting)

“So my plan is to use Dijkstra's — I'll keep a min-heap so I always grab the closest unvisited node first.”

Handling errors

Written (Slack/PR)

Edge case: disconnected graph nodes should return -1.

Spoken (Meeting)

“Oh, and if a node can't be reached from the source, I'll just return negative one for that.”

Example Dialogue

InterviewerGiven a weighted directed graph, find the shortest path from a source node to all other nodes.

YouBefore I start, let me clarify a few things. Can the edge weights be negative?

InterviewerNo, all weights are positive.

YouGreat — since all weights are positive, Dijkstra's algorithm is a good fit. Let me walk you through my approach.

YouI'll use a min-heap to always process the node with the smallest known distance. I'll maintain a distances array initialized to infinity, except the source which starts at zero.

InterviewerSounds good. What's the time complexity?

YouWith a binary min-heap, it's O of (V plus E) log V, where V is the number of vertices and E is the number of edges. The log V factor comes from the heap operations.

InterviewerCan you handle the case where a node is unreachable?

YouYes — any node that still has a distance of infinity after the algorithm finishes is unreachable from the source. I'll return negative one for those in the output.

YouLet me code this up and trace through a small example to make sure it works.

Common Questions

How do I think aloud during a technical interview in English?

Practice narrating your thought process: 'Let me start by understanding the constraints… Okay, so the input is a graph with positive weights… My first instinct is to use Dijkstra's because…' The key is to talk while thinking, not think silently and then talk. Practice this with Whisperly's AI coach.

What should I say when I'm stuck?

Be transparent: 'I'm going to take a moment to think about this.' Or ask for a hint diplomatically: 'Could you give me a small hint about the direction? I'm considering two approaches but I'm not sure which is more efficient.'

How do I pronounce Big O notation?

Say 'O of n' (for O(n)), 'O of n squared' (for O(n²)), 'O of n log n' (for O(n log n)), and 'O of one' (for O(1), constant time).

Stop Stumbling on Tech Calls

Practice explaining code, architecture, and bugs with an AI coach that understands engineering context.

Start Practicing Now

No credit card required.

More Tech Scenarios

System Design Discussions

Explaining Architecture Decisions

Code Review Discussions