The 4 Interview Rounds (And Why Most People Fail at #1)
A typical tech interview in 2026 has four stages. Most candidates obsess over coding challenges and ignore everything else. That's a mistake — more candidates are eliminated in behavioral rounds than coding rounds .
- Recruiter Screen (15-30 min) — Quick fit check: salary expectations, work authorization, timeline, and a "tell me about yourself" warm-up. 40% of candidates are eliminated here.
- Behavioral Interview (45-60 min) — Deep questions about past experience, conflicts, failures, and leadership. Uses the STAR framework. 35% elimination rate.
- Technical Rounds (2-4 hours) — Coding challenges, system design, and/or take-home projects depending on the company. 25-50% elimination rate.
- Culture Fit / Team Match (30-45 min) — Meet the team, ask questions, evaluate the environment. Rarely eliminates candidates but influences offer terms.
Let's break each one down.
Behavioral Interviews: The STAR Method Masterclass
"Candidates who ask me questions about team challenges and growth plans always stand out. It shows they are evaluating us too, not just hoping to be picked." — Priya S., Engineering Director
The STAR framework turns rambling stories into structured, impressive answers. Every behavioral question can be answered with this format:
- S — Situation: Set the scene in 1-2 sentences. "The authentication service was going down 3 times a week."
- T — Task: What was your specific responsibility? "I was tasked with improving uptime to 99.9%."
- A — Action: What did YOU do (not the team)? Be specific. "I implemented circuit breakers, added retry logic, and set up monitoring with PagerDuty."
- R — Result: Quantify the outcome. "Uptime improved to 99.97%. Overnight pages dropped from 12/week to 1/month."
The 7 Questions You WILL Be Asked
Prepare STAR stories for each of these — they cover 90% of behavioral interviews:
- "Tell me about yourself" — Not your life story. 2-minute pitch: current role → key achievement → why you're here. See: The "Tell Me About Yourself" Formula for Developers .
- "Tell me about a conflict with a coworker" — They want to see emotional intelligence, not who was right.
- "What's your biggest weakness?" — "Perfectionism" is dead. Name a real weakness + what you're doing about it. Deep-dive: How to Answer the Weakness Question .
- "Tell me about a failure" — Describe a real production incident or project failure. Emphasize what you learned. Read: How to Answer "Your Biggest Production Failure" .
- "Why are you leaving your current job?" — Never badmouth. Focus on growth, learning, and alignment with the new role.
- "Where do you see yourself in 5 years?" — Show ambition that aligns with the company's growth trajectory.
- "Do you have any questions for us?" — Always have 3-5. Ask about team challenges, technical decisions, or growth paths.
For the dark patterns in behavioral screening, read: The Behavioral Interview Maze: HR Screening's Shadow Zone .
System Design Interviews: The Framework
System design questions test your ability to architect scalable solutions. There's no single "right" answer, but there IS a framework that interviewers expect:
The 5-Step System Design Framework
- Clarify Requirements (3-5 min) — Ask: How many users? Read-heavy or write-heavy? Latency requirements? "Design Twitter" is a different answer for 1K users vs 500M.
- High-Level Design (5-8 min) — Draw the major components: load balancer, API gateway, application servers, database, cache, message queue. Don't go deep yet.
- Deep-Dive (10-15 min) — The interviewer will pick 1-2 components to explore. Database schema, caching strategy, API design — be ready to go deep on any component.
- Trade-offs (5 min) — Explain WHY you chose each technology. "I chose Redis for caching because of sub-millisecond reads, but DynamoDB for persistent storage because of automatic scaling."
- Scalability & Edge Cases (5 min) — How does the system handle 10x traffic? What happens when a component fails? How do you monitor it?
For API-specific design questions, our deep-dive covers REST vs GraphQL vs gRPC: Mastering the API Design Interview . For database sharding patterns: Database Sharding for System Design Interviews .
Coding Challenges: Patterns Over Problems
The biggest mistake in algorithm prep is solving 500 random LeetCode problems. Pattern recognition beats volume. There are approximately 15 core patterns that cover 90% of interview questions:
The 10 Most Important Patterns
- Two Pointers — Sorted arrays, palindromes, pair sums
- Sliding Window — Subarrays, substring problems, maximums within ranges
- Fast & Slow Pointers — Cycle detection, middle of linked list. Deep-dive: Mastering Fast and Slow Pointers
- BFS / DFS — Tree traversals, graph problems, island counting
- Dynamic Programming — Coin change, longest subsequence, knapsack
- Binary Search — Not just sorted arrays — search space reduction in any monotonic function
- Greedy Algorithms — Activity selection, interval scheduling. See: Proving Optimality in Greedy Algorithms
- Stack / Queue — Valid parentheses, next greater element, monotonic stack
- Hash Maps — Frequency counting, two-sum variants, grouping
- Backtracking — Permutations, combinations, N-Queens, Sudoku solver
For each pattern, solve 3-5 problems of increasing difficulty. Understand WHY the pattern applies, not just HOW to code it. For a curated list, read: The Hardcore 25: Algorithms, System Design, and LeetCode Mastery .
Concurrency: The Senior Differentiator
For senior roles, expect concurrency questions: deadlocks, race conditions, thread safety, and distributed locking. This is where mid-level engineers stumble. Our guide: Concurrency Explained: Deadlocks, Race Conditions, and More .
AI in Interviews: What You Need to Know
In 2026, AI is increasingly present in the interview process:
- AI video screening — Some companies (HireVue, Pymetrics) use AI to analyze facial expressions, speech patterns, and word choice during recorded interviews. This is controversial and banned in some jurisdictions.
- AI-assisted coding tests — Platforms track typing patterns, tab-switching, and clipboard usage to detect cheating or AI assistance.
- Cultural fit algorithms — Some companies score your behavioral interview transcripts using NLP models.
What to do: Be yourself, look at the camera (not the screen), speak clearly, and don't try to game the system. Authenticity scores better than performance. More details: How AI Recruiters Analyze Your Emotions in Video Interviews .
Salary Negotiation: The Data-Driven Approach
The negotiation starts the moment the recruiter asks "What are your salary expectations?" — which is usually in the first 5 minutes of the process. How to handle every stage:
Before the Interview: Research
- Levels.fyi — Verified compensation data for tech companies, broken down by level
- Glassdoor / Blind — Self-reported data (less accurate, wider coverage)
- Salary.com — Industry benchmarks by location and experience
During the Interview: Deflect
When asked about salary expectations, respond with: "I'd love to learn more about the role first. Can you share the range you've budgeted for this position?" In many states (California, Colorado, New York, Washington), companies are legally required to disclose salary ranges.
After the Offer: Negotiate
- Express enthusiasm first — "I'm excited about this opportunity and the team."
- Ask for time — "I'd like to review the details carefully. Can I get back to you by [date]?"
- Counter with data — "Based on Levels.fyi data for L5 engineers in [city], the range is $180K-$210K. Given my experience with [specific skill], I'd like to target $200K."
- Negotiate beyond base salary — RSUs/equity, signing bonus, start date, remote flexibility, and learning budget are all negotiable.
For remote salary negotiation, read: COLA: Mastering Remote Salary Negotiation .
Red Flags: How to Evaluate the Company Back
The interview is a two-way evaluation. Here are red flags that predict a problematic employer:
- No salary range disclosed — In jurisdictions that require it, this is a legal issue. Elsewhere, it suggests bad-faith negotiation.
- "We're like a family" — Often translates to: poor boundaries, expected overtime, guilt for using PTO.
- Vague role description — "Wear many hats" means undefined responsibilities and no promotion path.
- High interviewer turnover — If the person who interviewed you leaves before you start, it's a signal.
- Excessive interview rounds — More than 5 rounds suggests organizational dysfunction or a culture of distrust.
- Take-home projects > 4 hours — Disrespectful of candidate time. Companies that value engineers pay for their time.
The 30-Day Interview Prep Plan
If you have one month before interviews, here's how to allocate your time:
Week 1: Behavioral
Write STAR stories for the 7 core questions. Practice saying them out loud (not reading). Record yourself and review.
Week 2: System Design
Study 5 classic designs: URL shortener, chat system, newsfeed, search engine, payment system. Practice drawing on a whiteboard or tablet.
Week 3: Coding
Master 10 patterns, solving 3-5 problems per pattern. Focus on medium-difficulty problems, not easy or hard extremes.
Week 4: Mock Interviews + Salary Prep
Do 3+ mock interviews with friends or on platforms like Pramp/Interviewing.io. Research salary data for your target companies and levels.
Ghosted After the Interview? What to Do
Getting ghosted after a final interview is frustrating but common. Wait 5-7 business days, then send a polite follow-up. If no response after the second follow-up, move on — but document the experience for Glassdoor. For email templates: Follow-Up Email Scripts When You're Ghosted .