Academic snapshot
Marks: 60% or 6.0 CGPA throughout 10th, 12th, graduation
Degree: Any graduate/postgraduate (B.E./MBA/M.Sc/BBA common)
Backlogs: None active on most drives
Year: Final-year and recent grads (within ~2 years)
Mu Sigma Placement Papers 2026
Mu Sigma is a decision sciences and analytics firm (founded 2004) with Bengaluru as the India hub and a global HQ story in Chicago.
HQ: Bengaluru, India (Global HQ: Chicago, IL, USA)
Employees: 3,500+
Revenue: $200+ Million USD
Website: mu-sigma.com ↗
Explore: Online Assessment · Coding Questions · Interview Experience · Preparation Guide · Certifications & Career · Download PDF
Who can apply to Mu Sigma in 2026?
Section titled “Who can apply to Mu Sigma in 2026?”Mu Sigma is not a mass Java hirer. The printed floor is often 60% throughout, and non-engineering degrees are welcome. The real cut is puzzles and structured talk.
Who this hub is for
Section titled “Who this hub is for”- For: Students targeting Business Analyst / Trainee Decision Scientist from engineering, MBA, stats, economics, maths, or other full-time degrees.
- Not the same track: Pure software-product SWE loops (those need DSA OAs). Experienced consulting manager searches.
- Engineers: Eligible, but LeetCode alone will not save a silent puzzle round. Bring SQL anyway. The 2025 Noida panel still poked joins.
On-campus vs off-campus
Section titled “On-campus vs off-campus”| Route | What usually changes | What stays the same |
|---|---|---|
| Campus (including non-engg) | PPT, college list, sometimes paper-based | Puzzle + case bar |
| Careers / portals | Resume volume | Still expect a puzzle assessment |
| Referral | Visibility | You still think out loud on cases |
Off-campus does not mean easier Fermi problems. It often means more applicants who treat it like CAT quant and freeze when asked to estimate.
Degree, 6.0, and “is 6.0 enough?”
Section titled “Degree, 6.0, and “is 6.0 enough?””Is 6.0 CGPA enough? It matches the commonly reported floor. A 6.3 who talks through a bus-full-of-golf-balls estimate beats an 8.8 who waits for the “correct” integer. A 7.5 with 58% in 12th still dies if the JD says throughout.
Branches: Any listed graduate/postgraduate stream. Puzzle skill > branch prestige.
Gaps and backlogs: Active backlogs are a usual blocker. Gaps need an honest skills story.
What helps the resume screen: Anything that shows structured thinking (research project, analytics internship, even a well-told campus fest ops story). See certifications after puzzles, not instead of them.
Skills they actually filter
Written: Puzzles, estimation, light quant
Case: Structure under ambiguity
Analytics: SQL/stats on some panels
Tools: SQL, Excel; Python/R helpful
Detailed eligibility criteria breakdown
Section titled “Detailed eligibility criteria breakdown”| Item | Typical expectation (candidate-reported) | Reality check |
|---|---|---|
| Graduation | 60% or 6.0 | Mailer wins |
| 10th & 12th | Often same throughout rule | Keep scans |
| Degree type | Any full-time grad/PG | Engineering not mandatory |
| Backlogs | None active | Confirm drive |
| Coding | Not the main BA bar | SQL still shows up |
Who usually fits and who struggles
Section titled “Who usually fits and who struggles”Good fit: You like messy problems, you will talk while you think, and you can live with Bengaluru (or the city on the letter).
Harder path: You only grind DSA; you freeze until you know the answer; you treat cases like a memorised 3C framework dump; you want a pure SWE title.
Mu Sigma previous year paper PDFs
Section titled “Mu Sigma previous year paper PDFs”Download 2026 Mu Sigma placement papers
2026 pattern in detail
2025 placement papers
2024 placement papers
Previous year question papers
Mu Sigma does not publish one permanent official paper set. Papers on placementpapers.app are pattern trainers for puzzles, estimates, and cases.
| Round (typical) | What to practice |
|---|---|
| Written / OA | Puzzles, Fermi, probability, light DI |
| Case / analytical | Structure, assumptions, recommendation |
| Analytics / technical | SQL, stats, project |
| Video synthesis / bot (some) | Clear summary, resume facts |
| HR | Why decision sciences, city, integrity |
Themes that keep repeating (2024-2026)
Section titled “Themes that keep repeating (2024-2026)”- Classic puzzle OA: 60-90 min, 20-30 items, brain teasers and estimates weighted harder than CAT-style DI.
- Dec 2024 Bengaluru (GFG): Aptitude plus video synthesis (watch, analyse, summarise in a text editor) and a speaking prompt. 400+ of ~1100+ shortlisted; panel later revisited the video answers.
- Oct 2025 Noida, 2026 batch (GFG): 2-hour campus aptitude, 20-minute AI bot resume interview, 45-minute technical on projects, DSA, joins, SQL debugging. Funnel: 96/800 aptitude, 80 bot, 21 technical. Feedback: “good student,” stronger SQL needed.
- 2026 prep expectation: Puzzles still first. Add a clean written summary habit and SQL joins. Do not assume a pure paper test.
How to use one paper set in a single sitting
Section titled “How to use one paper set in a single sitting”- Timed: 60-90 minutes of puzzles/quant. Speak assumptions even if you write them.
- Untimed: rewrite one estimate with explicit steps (population × rate × time).
- One 20-minute mini-case on paper: clarify, split, recommend.
- Three SQL queries from memory.
- Watch a 5-minute product video and write a 150-word synthesis (for video-round hygiene).
Related drills
Section titled “Related drills”| Theme | Drill next | Why Mu Sigma asks it |
|---|---|---|
| Fermi estimate | Golf balls / buses / market size | Ambiguity |
| Probability puzzle | Conditional, Bayes lite | Decision under uncertainty |
| Case structure | Churn / pricing / store count | Client work |
| SQL joins | Debug a wrong join | 2025 technical notes |
| Video synthesis | 3 insights + one risk | 2024 OA extra |
Mu Sigma placement process, round by round
Section titled “Mu Sigma placement process, round by round”Mu Sigma hires through campus cells (including MBA and sciences), careers / portals, and referrals. After eligibility, the shape is analytics-consulting: puzzle OA, case, analytics/SQL, HR. Extras: video synthesis, AI bot.
Application methods
Section titled “Application methods”Campus recruitment
Engineering, MBA, stats, economics campuses. Listen for video/bot language in the PPT.
Online applications
mu-sigma.com careers or listed portals. Follow the invite for format.
Referrals
Improve screen odds. Cases still require thinking out loud.
1. Resume / eligibility screen
Section titled “1. Resume / eligibility screen”60% / 6.0 throughout when printed, any listed degree, no active backlogs. Gate.
2. Written / Online Assessment (60-90 min, sometimes ~2 hours)
Section titled “2. Written / Online Assessment (60-90 min, sometimes ~2 hours)”Student-reported working map:
| Section (typical) | Questions (reported range) | Time feel | Focus |
|---|---|---|---|
| Puzzles & logical | 10-15 | 30-40 min | Teasers, patterns, estimation |
| Quantitative | 8-10 | 20-25 min | Arithmetic, probability, DI |
| Case / written (some) | 1 | 15-20 min | Structure, not a “correct” essay |
Pass habit: skip a stuck riddle in 90 seconds. Write assumptions. Negative marking varies. Student rumours put written pass around 20-30%. 2025 Noida: 96/800 after a 2-hour aptitude. Your invite wins.
Video synthesis extra: watch, note 3 facts, write a clean summary. The Dec 2024 panel reused those answers. Do not write a poem.
AI bot extra: 20-minute resume interview. Short, consistent facts. Do not ramble into a fake internship.
3. Case / analytical interview (45-60 min)
Section titled “3. Case / analytical interview (45-60 min)”Open business problem (churn, pricing, store footprint). Data when given. Follow-up puzzles. They grade the path.
Pass habit: clarify the goal, split into 3 buckets, estimate one bucket, give a recommendation and a risk. “I need a week of data” is not a close.
4. Technical / analytics (30-45 min, role-dependent)
Section titled “4. Technical / analytics (30-45 min, role-dependent)”SQL, stats, project. 2025 Noida also poked DSA and join debugging. If you are a non-coder, still know SELECT / JOIN / GROUP BY.
5. HR (~30 min)
Section titled “5. HR (~30 min)”Why Mu Sigma, why decision sciences (not “I like data”), city, joining, integrity. Know the ₹6-9 LPA trainee band (some notes say ₹5-8). The letter wins.
Timeline from application to offer
Section titled “Timeline from application to offer”| Phase | Typical duration | Notes |
|---|---|---|
| Apply / screen | 3-5 days | Degree + % check |
| OA (+ video/bot) | 1 sitting to a few days | Formats vary |
| Case + analytics | 1-2 days | Virtual common |
| HR | Same day or next | Short |
| Offer | 3-5 days | Written letter wins |
| End to end | ~2-3 weeks campus often | Off-campus can stretch |
Fermi habit (the estimate people skip)
Section titled “Fermi habit (the estimate people skip)”Pick a number, state the assumption, multiply, sanity-check against a known fact (India population, a cricket stadium, a city block). Interviewers want the chain. A silent “um, 10,000?” is the fail.
Example shape: golf balls in a bus → bus volume as box × packing fraction × ball volume. Say 70% packing, not 100%. Stop at one significant digit unless they ask you to refine.
Case habit (12 minutes on paper)
Section titled “Case habit (12 minutes on paper)”- Goal in one sentence.
- Three drivers.
- One calculation.
- Recommendation + one risk.
- What data you would collect next week.
That beats a memorised framework dump.
Weekend mock
Section titled “Weekend mock”Saturday 10:00: 75-minute puzzle set.
Saturday 12:00: one Fermi rewrite.
Saturday 16:00: 20-minute case with a friend who interrupts.
Sunday 11:00: 5 SQL queries + one join debug.
Sunday 12:00: 150-word video synthesis.
Sunday 12:30: HR why-decision-sciences.
Two weekends of that beat a riddle PDF binge.
Documents pack
Section titled “Documents pack”10th, 12th, degree, ID, resume one page. Throughout 60% means the 12th PDF matters.
How interviews get scheduled
Section titled “How interviews get scheduled”Campus: PPT → long aptitude (60-120 min) → optional video/bot → case → analytics → HR. Off-campus: similar, with more resume noise. Screenshot the invite. If the PPT never mentioned an AI bot, do not panic-build a bot course. If it did, record your intro once and make the resume boringly true.
After a puzzle mock, the only review that matters
Section titled “After a puzzle mock, the only review that matters”Pick the three items you skipped or guessed. Rewrite the assumption chain. Speak one of them to a wall. That is the whole review. Re-reading 40 riddle answers you already got right is how people feel busy and stay silent in the cabin.
What’s on the Mu Sigma test & where to practice
Section titled “What’s on the Mu Sigma test & where to practice”Online assessment
OA format, timing, and puzzle types.
Coding questions
Light DSA/SQL practice for panels that still poke it.
Aptitude / mock quiz
Puzzles, quant, and timed mock quiz.
Interview experiences
Candidate stories and round notes.
HR interview questions
Common HR questions, sample answers, STAR tips.
Week-by-week prep
Study plan and round-by-round strategy.
A practical Mu Sigma prep plan
Section titled “A practical Mu Sigma prep plan”If a Mu Sigma invite is 3-8 weeks out, treat this hub as the map and the nested pages as the drills. Do not reread eligibility ten times. Check the JD once, then spend hours on the filter that actually rejects people.
Typical fresher CTC chatter for Mu Sigma sits around ₹6-9 LPA (candidate-reported - confirm the letter). Languages students mention most often: Java, Python, C, Go. Pick one and stay with it in OA and interviews.
Eligibility reminder (from student reports): Candidate-reported floor is often 60% or 6.0 CGPA across 10th, 12th, and graduation. Any graduate/postgraduate degree is commonly accepted; engineering is not mandatory. Final-year or recent grads; no active backlogs on most drives. Puzzle and case performance matter more than branch. Verify the notice.
What you are training for: Typical loop: written/online assessment (60-90 min, puzzle-heavy) → case/analytical interview (45-60 min) → technical/analytics interview (30-45 min, role-dependent: SQL, stats, projects) → HR (~30 min). Some 2024-2025 notes add video synthesis or an AI-bot resume screen. About 2-3 weeks on many campus batches.
Short version of the prep split: Split time roughly 40% puzzles and estimation (think out loud), 30% case structuring, 20% stats/SQL/DI, 10% HR and ‘why decision sciences.’ Coding is not the main bar for most BA roles. Full plan: placementpapers.app/musigma/preparation-guide/.
A four-week plan that actually gets used
Section titled “A four-week plan that actually gets used”Week 1 - pattern, not vibes. Oa pattern first: sit one untimed paper, then redo it under the real clock. Open the Mu Sigma previous-year sets on this site and mark which sections ate your time. Write the timing on a sticky note. That number is more useful than a 200-problem LeetCode streak you never timed.
Week 2 - the steep filter. Most Mu Sigma drives fail people on the first timed paper or OA. Do three timed sits this week. After each one, log every miss in a two-column note: topic and why (slow, wrong formula, bad edge case, panic). The next sit should only add problems from that log.
Week 3 - talk tracks. Technical interviews at Mu Sigma usually want a clean project story, one or two CS fundamentals, and (if the process has it) a behavioral or values round. Write STAR stories on paper, not in your head. Practise “Why Mu Sigma?” with one concrete product, lab, or business line - not a slogan.
Week 4 - mocks and logistics. Star stories + one project walkthrough recorded on your phone. If you still have energy, redo only the questions you failed in week 2.
Where the hours should go
Section titled “Where the hours should go”| Slice | Share of weekly hours | What “done” looks like |
|---|---|---|
| Timed papers / OA | ~40% | You finish a Mu Sigma-style sit without guessing the clock |
| Coding / role technical | ~30% | You can explain a solution out loud in your chosen language |
| Fundamentals / domain | ~20% | You can teach one OS/DBMS/networks (or domain) topic to a friend |
| Stories + HR | ~10% | Two STAR stories and a specific Why-this-company answer |
If you only have 10 days, keep the same ratios and cut volume, not the timed sits.
Nested pages to open this week
Section titled “Nested pages to open this week”- OA / assessment notes
- aptitude mock quiz
- coding question bank
- interview experiences
- HR question bank
- week-by-week guide
- previous year papers
- 2026 paper page
Use this hub for the Mu Sigma story (eligibility, rounds, salary). Use those pages for reps. If a card 404s on your campus mirror, the explore links at the top of this page are the same destinations.
Common ways students waste a month
Section titled “Common ways students waste a month”- Collecting 12 Mu Sigma PDFs and never sitting one under a timer.
- Switching languages every weekend because a senior used a different one.
- Memorising HR answers before they can finish the OA.
- Ignoring the company-specific quirk on this page (unique round, exam name, or role band) and preparing like a generic service-company drive.
If you do one thing after reading this: schedule three timed Mu Sigma papers on three different days, then interview prep on the leftovers.
If you are starting from a service-company prep base
Section titled “If you are starting from a service-company prep base”A TCS/Infosys aptitude deck still helps the quant/logic slice of many Mu Sigma papers, but do not assume the same cutoff culture. Read the process section on this page once, then swap in Mu Sigma-specific coding or domain drills from the nested banks. Shared prep is fine; shared assumptions about rounds are not.
10-day crash version
Section titled “10-day crash version”Days 1-3: one timed Mu Sigma paper each day plus error log. Days 4-6: only the topics you missed, plus 4-6 coding problems in one language. Days 7-8: project + STAR + Why Mu Sigma. Day 9: one more full mock. Day 10: light review and sleep. This is worse than four weeks, but it beats rereading notes the night before.
How to train for Mu Sigma puzzles and cases
Section titled “How to train for Mu Sigma puzzles and cases”Week calendars live on the preparation guide. This H2 is the focus split.
Time split
Section titled “Time split”| Bucket | Share | What “done” looks like |
|---|---|---|
| Puzzles + Fermi | ~40% | Timed sets, assumptions written |
| Cases | ~30% | 45-minute mock with a spoken close |
| SQL / stats / DI | ~20% | Joins you can debug |
| HR + video/bot hygiene | ~10% | 150-word summary + why-Mu-Sigma |
If you are an engineer who only codes, stop LeetCode for two weeks and talk through estimates. If you are an MBA who only cases, add SQL so a 2025-style panel does not strand you.
This week
Section titled “This week”- One timed puzzle mock from aptitude questions.
- Two Fermi problems with written assumptions.
- One case with a friend who says “so what?”
- Five SQL queries including one broken join.
- One video synthesis paragraph.
What not to waste time on
Section titled “What not to waste time on”- Memorising 200 riddle answers without talking
- Staff-level ML courses for a trainee BA drive
- Ignoring SQL because “Mu Sigma is not coding”
- Quoting 18 LPA in HR
- Treating the AI bot as a joke
Two-week focus if the drive is soon
Section titled “Two-week focus if the drive is soon”Week A: daily 45-minute puzzle timed sets; two Fermi write-ups; one 20-minute case.
Week B: SQL joins every morning; one full 60-minute case with a friend; one video synthesis; HR why-decision-sciences recorded twice.
Engineers: cut DSA to maintenance (3 problems/week) until puzzles are not silent. Non-engineers: do not skip the five SQL queries even if the PPT never said “coding.”
How to say “why Mu Sigma” without a brochure
Section titled “How to say “why Mu Sigma” without a brochure”Decision sciences is structuring messy client problems with data, not building a consumer app. You like estimates and trade-offs. You will relocate to Bengaluru (or the city on the letter). You know the trainee CTC band and you will read the letter. That is enough. “I have always been passionate about leveraging analytics” is how people sound like a template.
Puzzle types worth a week, not a lifetime
Section titled “Puzzle types worth a week, not a lifetime”You do not need 400 riddles. You need representatives:
- Pattern / series that break if you only look at first differences
- River / weighing / truth-liar where the win is stating the information update
- Probability with a sample space you draw
- Fermi with a written chain
- Geometry lite (how many, not JEE)
For each type, do three timed, then explain one to a roommate in 90 seconds. If you cannot explain it, you do not own it. Mu Sigma interviewers will ask a cousin, not the same joke from a PDF.
SQL debug they actually used
Section titled “SQL debug they actually used”The 2025 Noida panel mentioned SQL debugging. Practise the boring failures:
JOINon the wrong key (duplicate rows)LEFT JOINthenWHEREon the right table (turns it into inner)GROUP BYmissing a columnCOUNT(*)vsCOUNT(col)with nulls
Write the wrong query, then the fix, aloud. That is more useful than a window-function course the night before.
Funnel math: 96 → 80 → 21
Section titled “Funnel math: 96 → 80 → 21”800 sat aptitude. 96 passed. 80 faced the bot. 21 reached technical. You cannot charm the 800. You can make sure a bot-friendly resume and five joins exist before the 96 becomes 21 without you. If you are in the 21, the panel already thinks you can think. Do not waste it by going silent on a join.
Night-before checklist
Section titled “Night-before checklist”- Charger, ID, resume that matches what a bot might ask
- One Fermi and one case close on paper
- Five SQL queries including a broken join
- 150-word synthesis practice if video was named
- Sleep. Silent panic on a morning puzzle is how 800 becomes 96 without you
What decision-sciences hiring means
Section titled “What decision-sciences hiring means”Mu Sigma is not trying to be TCS and not trying to be Google. The puzzle brand is real. The 2025 SQL panel is also real. You can be a stats graduate or a CSE graduate; both still have to talk. You cannot outsource the think-out-loud habit to a dump PDF. You cannot skip joins because a senior said “they only do puzzles.” Read your mailer. Then practise both.
If Mu Sigma, Fractal, and EXL all land the same week, compare how heavy SQL vs puzzles felt in recent notes, city, and the letter. Do not pick on brand memory from 2014 campus stories.
Mu Sigma fresher CTC in India (2026)
Section titled “Mu Sigma fresher CTC in India (2026)”Figures below are candidate-reported India bands (as of August 2026).
| Track | Typical CTC (candidate-reported) | Notes |
|---|---|---|
| BA / Trainee Decision Scientist | ₹6-9 LPA | Primary band; some notes ₹5-8 |
| BA (1-2 years) | ₹9-12 LPA | Not campus default |
| Senior BA (indicative) | ₹12-16 LPA | Experience |
| Data engineer (entry, some posts) | ₹7-10 LPA | SQL/pipeline heavier |
Confirm the written offer. Training and variable language varies.
Benefits students actually care about
Section titled “Benefits students actually care about”Health insurance, structured decision-sciences training, hybrid on some teams, Bengaluru (and other) city norms. Ask HR about joining location. Do not assume Chicago.
How to talk CTC in HR
Section titled “How to talk CTC in HR”Know ₹6-9 LPA (and that some batches hear ₹5-8). Ask how the package splits. One calm question beats a consulting-firm 20 LPA demand.
Offer-stage discipline
Section titled “Offer-stage discipline”- Read city
- 10th/12th/degree % must match
- Bot/video shortlist is not an offer PDF
Recent Mu Sigma placement trends
Section titled “Recent Mu Sigma placement trends”- Puzzles remain the brand. Cases have more weight than a decade ago on many drives.
- Video synthesis appeared in 2024 Bengaluru notes; treat written summary as a real skill.
- AI bot + SQL panel appeared in 2025 Noida notes; engineers cannot skip joins, non-engineers cannot skip puzzles.
- ₹6-9 LPA is the commonly cited trainee band; confirm the letter.
- Hiring still mixes engineering and non-engineering campuses.
Net for 2026: do not prepare as if this were TCS NQT or a product DSA OA. Think out loud. Keep SQL warm enough to survive a 21-person technical funnel.
Bengaluru vs Noida (and why the letter wins)
Section titled “Bengaluru vs Noida (and why the letter wins)”Bengaluru is the story hub. The 2025 batch note was Noida. Do not memorise one city. Ask HR. If you cannot go, say so before you celebrate a 21-person shortlist. Relocation is not a puzzle. It is a yes/no that should already be written on paper, the same way a Fermi assumption should be written before you multiply. Students who “will decide after the offer” are the ones who stall joining and annoy the next campus cycle. Decide the city on paper this week, before the bot round, not after the PDF. Write the yes/no next to your Fermi assumptions so both are already decided before anyone asks, including the AI bot if that extra screening round shows up on your official campus invite.
FAQs students ask about Mu Sigma placements
Section titled “FAQs students ask about Mu Sigma placements”Papers & process
Section titled “Papers & process”What is the Mu Sigma placement process for freshers?
Typical path: puzzle-heavy OA → case interview → analytics/SQL (role-dependent) → HR. Some drives add video synthesis or an AI bot. About 2-3 weeks on many batches.
How many rounds does Mu Sigma usually have?
Most fresher loops report 3-4 after resume. Count the mailer.
What is the exam pattern?
Student-reported 60-90 minutes (sometimes ~2 hours): puzzles first, then quant/DI, sometimes a written case. Details: online assessment.
Eligibility & offer
Section titled “Eligibility & offer”What eligibility / CGPA do fresher drives usually expect?
Commonly 60% or 6.0 throughout; any listed graduate/PG degree; no active backlogs. Your notice wins.
What is Mu Sigma fresher salary in India?
Candidate-reported trainee CTC often ₹6-9 LPA (as of August 2026; some notes ₹5-8). The written offer is the only number that matters.
Prep & Mu Sigma-specific
Section titled “Prep & Mu Sigma-specific”How should I prepare?
Roughly 40% puzzles/Fermi, 30% cases, 20% SQL/stats, 10% HR/video hygiene. Guide: preparation guide.
Do I need coding skills?
Not for most BA/Decision Scientist drives. SQL still appears on some 2025 panels. Do not skip puzzles for LeetCode.
Why the puzzle interviews?
Decision sciences work is ambiguous. They watch how you structure, not whether you memorised a riddle.
What is video synthesis or the AI bot?
2024 Bengaluru: watch/analyse/summarise plus speaking. 2025 Noida: 20-minute AI-bot resume screen before a SQL-heavy technical. Both are real cuts when present.
Mu Sigma vs Fractal vs EXL?
Mu Sigma is puzzle-and-case first with mixed degrees. Fractal often leans more ML/tech earlier. EXL mixes analytics with ops/insurance flavour. Compare recent write-ups and letters.
Any pro tip that actually changes outcomes?
Think out loud. State assumptions. Keep SQL joins warm. Do not wait for the perfect answer.
Where can I read real interview experiences?
Mu Sigma interview experience plus the GFG notes in the stories section.
Can non-engineers apply?
Yes on most BA/Decision Scientist drives. Puzzle and case performance matter more than a B.Tech label.
Should engineers skip puzzles and only do SQL?
No. Puzzles still open the funnel (96/800 in one 2025 note). SQL can strand you later. Do both at fresher depth.
Mu Sigma - recent candidate experiences
Section titled “Mu Sigma - recent candidate experiences”Trainee Decision Scientist, Bengaluru (GFG, updated Dec 2024)
Section titled “Trainee Decision Scientist, Bengaluru (GFG, updated Dec 2024)”Unusual OA: aptitude plus a Video Synthesis task (watch, analyse, summarise in a text editor) and a speaking prompt. “Out of approximately 1100+ candidates, 400+ students were shortlisted for interviews,” where the panel revisited the video answers before projects and HR. Source: GeeksforGeeks Mu Sigma Bengaluru.
Takeaway: written summary quality is part of the loop. Do not treat the video task as a break.
Trainee Decision Scientist, Noida, 2026 batch (GFG, 27 Oct 2025)
Section titled “Trainee Decision Scientist, Noida, 2026 batch (GFG, 27 Oct 2025)”2-hour campus aptitude, 20-minute AI bot resume interview, 45-minute technical on projects, DSA, joins, SQL debugging. Funnel: “Out of 800 students, 96 cleared aptitude, 80 faced bot interview, and 21 reached the technical panel.” Feedback praised being “a good student” while urging stronger SQL; final selection followed all rounds. Source: GeeksforGeeks Mu Sigma 2026 batch.
Takeaway: puzzles get you in; SQL can still strand you at 21/800. Bot interviews punish messy resume stories.
Composite notes students repeated
Section titled “Composite notes students repeated”What helped: thinking out loud; Fermi with stated assumptions; one clean case close; joins from memory; a why-decision-sciences answer that is not “I like Excel.”
What hurt: silent riddle staring; framework name-dropping; fake internships the bot can trip; ignoring SQL as a non-coder; quoting product-company CTC.
After you clear OA
Section titled “After you clear OA”- Re-work every stuck puzzle with a spoken chain.
- One case mock with a friend who interrupts.
- Five SQL queries including a broken join.
- Re-read your video synthesis if you wrote one.
- Sleep and city answer. Do not open a new ML course.
Video synthesis rubric (from the 2024 note)
Section titled “Video synthesis rubric (from the 2024 note)”Watch once without notes. Watch again and capture: what happened, three facts, one implication, one risk. Write 120-180 words in clean sentences. No slang. No “in conclusion.” The panel may ask you to defend a sentence you wrote. If you cannot, you copied a vibe instead of watching.
AI bot hygiene (from the 2025 note)
Section titled “AI bot hygiene (from the 2025 note)”- Resume bullets must be true and match what you will say to humans
- Short answers: role, stack, metric, what you would redo
- Do not invent a Mu Sigma internship
- If the bot asks a puzzle, think out loud in the same voice you will use in the case round
Mini case you should be able to run cold
Section titled “Mini case you should be able to run cold”Retailer wants to cut churn 10% in 90 days.
- Clarify: which segment, online vs store, what “churn” means.
- Buckets: price, product, service, habit/competition.
- Estimate: 1 lakh active, 5% monthly churn, 10% of that is service-fixable.
- Recommend: one service experiment + one win-back offer, measured weekly.
- Risk: discounting your way into worse margins.
That is a trainee-level close. A 12-slide McKinsey pastiche is not.
More narratives: Mu Sigma interview experience.
If you’re also targeting…
Section titled “If you’re also targeting…”Fractal · EXL · Tiger Analytics · ZS Associates · Genpact · Deloitte