Online assessments, aptitude tests, internships, graduate schemes and campus hiring: the rounds students and new graduates face before their first job, wherever they apply.
38 published across 7 topics.
Timed coding assessments, AI-assisted sections, hidden test cases, proctoring rules and when a retake is worth it.
An AI-assisted assessment section scores whether you understand the code before you change it, whether you verify every change by running the tests, and whether you can explain what you changed and why. The assistant speeds up reading and typing, but judgement is yours, and clever prompting earns nothing by itself.
Use code tracing, not reading. Write a small table with one column per variable and one row per step, count loop iterations from the bounds first, and apply the language's own rules for integer division, references and evaluation order. Skim the options to see which rule the question is testing.
Spend the first few minutes reading all four problems and their constraints, then solve them easiest first so points are banked early. Give each a time box, submit a working version once it passes your own checks, and take partial credit on hidden tests rather than chase one perfect answer.
Turn each constraint into a test: empty input, one element, the maximum size, duplicates, negatives, totals that overflow a 32-bit integer and inputs big enough to time out. Then compare your solution with a brute force on small random inputs; the first input where they disagree is your missing hidden test.
Proctored assessments usually require a verified ID, a webcam showing your face, a quiet room with nobody else present, one screen and only allowed materials. Read the rules before the day, set up so your eyes stay on the screen, and if anything goes wrong, note it and contact the support channel in your invitation.
Retake an online assessment only after you know which section cost you and have practised that section under the same timer. First check the invitation or ask the recruiter whether a new score replaces the old one and whether other employers can see either attempt.
Numerical, logical and verbal reasoning, critical-thinking tests and situational judgement, solved fast without guessing.
A situational judgement test compares your choices with a key built from how effective experienced people in the role judge each response. The heroic option, fixing everything alone and telling nobody, usually ranks low because it hides a risk from its owner. Near-miss rankings often earn partial credit.
In numerical reasoning tests, use multipliers for percentages so successive changes combine by multiplying, separate percentage change from percentage points, scale ratios to a common term, and estimate before calculating so the rough figure rules out wrong options and checks your exact one.
Decide first whether order matters: permutations if it does, combinations if it does not. Divide out repeated items, and for "at least one" count the complement and subtract. Picking one item first and then the rest freely double counts, because each outcome is reached several ways.
Solve syllogisms by drawing the weakest diagram the statements allow and accepting a conclusion only if it holds in every diagram, remembering that "some" means at least one. Solve a seating arrangement by placing fixed clues first, branching on the rest and eliminating, never stopping at the first layout that fits.
Yes. Time, speed and distance and work-rate problems are the same equation, amount = rate × time. Rates add when things work together or move towards each other and subtract when they oppose; for work, set the job to the LCM of the times so each rate is a whole number of units.
In a critical-thinking test, true means the passage guarantees the statement and false means the passage contradicts it. Everything else is cannot say, including statements that are plausible, likely or true in real life. Most wrong answers come from filling the gap with outside knowledge or an unstated cause.
First- and second-year internships, co-op and placement-year interviews, intern coding rounds and earning a return offer.
A co-op interview or placement year interview hires someone to join a real team for anywhere from four months to a year or more, so it probes whether you can become productive in an unfamiliar codebase, stay engaged across the whole term and work inside a team, not only whether you can learn quickly over a summer.
A first- or second-year internship interview expects core data structures from your first courses, arrays, strings, hash maps and simple loops, used correctly and explained aloud. Beyond that it scores how you reason, how you take a hint and how quickly you learn, not prior jobs or advanced algorithms.
Pick one coursework project, say plainly that it was an assignment and what the brief fixed, then spend the answer on the decisions you made inside that brief, the problem that cost you the most time and what you would change. Owned decisions and honest scope matter more than how impressive the project sounds.
A return offer mostly rests on what your manager and mentor can write about you: a project finished to an agreed scope, evidence you acted on feedback, and work they saw progress. Ask for explicit feedback at the midpoint, cut scope to finish, and write the work up. Headcount you cannot control also matters.
Coding without a compiler means doing the computer's checking yourself. Write two or three test cases before the code, keep functions small with clear names, then trace one test by hand line by line, aloud. The trace finds the bugs a run would have found, and it shows the interviewer how you verify.
Group exercises, case studies, strengths-based interviews, recorded video interviews and presentation exercises.
In an assessment centre group exercise, each assessor usually watches a few candidates and records observable behaviours: building on others' points, using facts from the brief, keeping to time, bringing quieter people in and moving to a decision. Airtime earns nothing; what you add when you speak does.
Treat a case study exercise as a timed decision, not a reading test. Read the request first, triage the pack by relevance and date, pull key figures into one table, then commit to one recommendation with reasons from the documents, the main risk and its mitigation, and minutes left to check the arithmetic.
Use the thirty seconds to choose one example and your first sentence, not to write a script. Spend the two minutes on a headline, a short situation, mostly what you did, and a result. Practise against a timer, test camera, light and sound first, and read the invitation for the retake rules.
A strengths-based interview listens for what you are good at and enjoy, shown in specific, recent examples told with natural detail and energy. There is no right strength; assessors check whether your preferences fit the role's daily work, so generic claims and scripted answers to competency questions score poorly.
Build one message, three supporting points and a timed close, not a full deck. Spend about a third of the preparation deciding what you think, a third building a few slides or a flipchart, and the rest rehearsing aloud. Aim to finish at nine minutes, because the questions afterwards are scored too.
Answer with the programme's structure, not the company's reputation. Name a specific rotation or training element, connect it to something you have already done, and say what you want to be able to do by the end. One accurate detail about the scheme beats three compliments about the brand.
Campus drives, career fairs and off-campus routes, new-grad coding rounds, and comparing first offers.
Expect easy-to-medium problems on arrays, strings, hashing, sorting, recursion and basic trees or graphs, graded as much on clean code and clear reasoning as on the algorithm. Calibrate by timing yourself on unseen mediums, out loud, rather than by counting solved problems.
Read the bond first and turn it into a number: what you would repay, over what period, if you left early. Then compare fixed first-year pay rather than headline salary, what the training and the startup's senior engineers would each teach you, and how stable each job is. Neither offer wins by default.
Through off-campus hiring routes: career fairs, graduate job boards and employers' early-careers pages, direct applications to smaller companies, and referrals from alumni and people you meet. Apply early in the hiring cycle, target roles properly, and give every application a project worth asking about.
Treat a campus placement drive as a series of separate elimination gates. The aptitude test checks speed and accuracy, the technical round checks fundamentals and your own resume, and HR checks communication and flexibility. Patterns and cut-offs vary, so take the official details from your placement cell, not forums.
Sort the five rejections by the stage where each one ended, because a pattern by stage tells you more than any single rejection. Then fix that stage only: ask for feedback where you can, run mock interviews that recreate it, and practise against it before the next process.
Explaining your own projects, a one-page resume with no work experience, and group or hackathon work you can defend.
Walk through your final-year project as problem, your part, two or three decisions with their trade-offs, what failed and what you would redo. The interviewer digs into whichever decision or result you cannot explain, so prepare the why behind every choice and the method behind every number.
Take one tutorial project somewhere the tutorial did not go, so you face a real decision on your own. Credit the tutorial in a README that says what you changed and why, add tests around your own code, and let the commits show the work. That gives your GitHub portfolio something to question.
Name the team size and your role, then the specific part of the hackathon project you built and one decision you made under the time limit. Say what in the demo was real and what was mocked or hard-coded. Your individual contribution, honestly scoped, is what the interviewer is scoring, not the prize.
A graduate resume with no work experience leads with education and projects written as decisions and outcomes, adds part-time work, and keeps a short skills list you can defend. Conventions differ by country: US resumes and UK CVs usually omit photos and personal details; some other markets more often include them.
Say plainly what your part was, why it was smaller, and what you would do differently, without inflating it or blaming teammates. A small, owned contribution described precisely beats a large one that collapses under two follow-up questions, and the lesson you drew is what gets scored.
Why hire you over other graduates, relocation and bonds, a first salary with nothing to anchor on, and explaining gaps.
Explain a gap year or repeated year in three sentences: what happened, what you did or changed, and the evidence that it is behind you. Keep it brief and honest, match your transcripts, and stop. Over-justifying an education gap is what turns an ordinary fact into a concern.
Anchor on market data for the role and city, not on a past salary you do not have. First check whether the offer is a fixed graduate rate or a band. A fixed rate rarely moves, so ask about start date, location or relocation help. A band can move with one polite, evidence-based ask.
A graduate's career goals answer should describe a learning path, not a title: what you want to be competent at after the first year, what you want to own by year three, and which direction you have not decided yet and how you will decide. Honest uncertainty scores better than a promotion timetable.
Do not compare yourself to people you have never met. Pick two or three things the role needs, give one piece of evidence for each from your projects, work or study, and end with why this role fits you. Evidence the panel can check beats adjectives every other graduate also uses.
Give a conditional yes where you mean it, ask for the specifics before you commit, and say no now to anything you cannot do. For a service agreement, ask for the duration, the repayment amount and what triggers it, and read the contract before signing. A yes you later retract costs more than an honest no.