Startup vs MNC Hiring: The Two Completely Different Evaluation Systems and How to Prepare for Both

The Indian entry-level tech hiring market is not one market. It is two parallel markets with entirely different evaluation criteria, interview formats, and decision-making processes. A candidate who prepared optimally for the TCS campus drive will fail a Razorpay interview in the first five minutes. A candidate whose GitHub portfolio impressed a startup CTO will fail the TCS NQT because they never practiced train-speed problems. The tragedy of tier-3 placement preparation is that placement cells prepare students for the MNC track exclusively, while the higher-compensation, faster-growth startup track operates on a completely different evaluation system that the placement cell never teaches. Understanding both systems, and preparing for both simultaneously, is the only way to keep your options open in a market where neither track guarantees an outcome.

Stage-by-Stage Comparison

Stage 1: Application screening. MNC: Your resume enters an ATS that filters on degree, percentage cutoffs (60% throughout academics for most service companies), and keyword matches. A human does not see your resume at this stage. Startup: Your application goes to either a recruiter who spends 30 seconds on your GitHub profile, or directly to an engineering manager who reviews your pinned repositories. Portfolio presence determines whether you advance. No portfolio = no advancement, regardless of academics.

Stage 2: First-round evaluation. MNC: Written aptitude test — quantitative, logical reasoning, verbal ability. Timed. Multiple choice. No coding. No portfolio review. The test score determines whether you proceed. Startup: Take-home assignment or live coding round. You are asked to build a small feature, fix a bug in an existing codebase, or design a simple API. Your code is evaluated on correctness, structure, error handling, and test coverage. Aptitude is never tested.

Stage 3: Technical interview. MNC: 20–30 minutes. Standard questions: "What is OOP?", "Explain the OSI model," "Write a program to check if a string is palindrome." The interviewer follows a script. Your ability to recite definitions matters more than your ability to reason about systems. Startup: 45–60 minutes. You discuss a project from your portfolio. The interviewer asks: "Why did you choose this database schema? What happens if this API endpoint receives 10,000 requests simultaneously? How would you add a caching layer to reduce response time?" Your reasoning ability matters more than your recall ability.

How to Prepare for Both Tracks Simultaneously

The preparation strategies for the two tracks are not mutually exclusive. They overlap in specific areas and diverge in others. Here is the allocation that covers both:

Shared preparation (serves both tracks): Build one deployed full-stack project with a relational database, visible commit history, and a live URL. This is the core asset that serves startup hiring directly and differentiates you in MNC interviews where most candidates have no portfolio. Prepare SQL — JOINs, GROUP BY, window functions. SQL is tested in both tracks. Prepare coding fundamentals — arrays, strings, hash maps, basic algorithms. These appear in both MNC compiler-based tests and startup live coding rounds.

MNC-specific preparation: Quantitative aptitude (train-speed, profit-loss, percentages, ratios). Logical reasoning (seating arrangements, blood relations, syllogisms). Verbal ability (reading comprehension, sentence correction). Allocate 25% of your preparation time to these if you are targeting both tracks. Allocate 50% if your portfolio is not yet ready and you are depending more heavily on MNC placements.

Startup-specific preparation: System design basics (URL shortener, rate limiter, notification system). Practice explaining your project architecture in 5 minutes — the "walk me through your project" question is the first question in 90% of startup interviews. Prepare for behavioral questions: "Tell me about a bug that took you a long time to fix. How did you debug it? What did you learn?" Startups evaluate learning velocity as heavily as current knowledge.