G-Research · Live Interview
G-Research Interview Questions & Prep
G-Research's first-round live interview is where strong written applications become offer pipelines or go nowhere. Below: the real questions G-Research asks, what they're testing for, and how to practise live until it feels routine.
Practise for G-Research
Start practising on Intervyo. Free tools, scored feedback, no payment.
- CV Checker, scored against G-Research
- HireVue practice, AI-scored
- Live AI mock interviews with Vyo
- Psychometric tests in real formats
- Application Tracker
The format
What G-Research's live interview actually looks like
Where in the timeline it sits: Candidate reports suggest that the live first-round interview follows immediately after the initial online screening stage, which typically consists of an automated quantitative aptitude or machine learning assessment.
Format
Live conversation, phone or video. 30 to 45 minutes. One-on-one with a manager or senior associate.
Interviewers
Manager or senior associate from your target team.
Structure
Interviewers probe weak answers. The candidates who get through handle follow-ups confidently, not just the headline question.
Real questions
25 real G-Research interview questions
Pulled from past applicants. These come up regularly in the live round.
- 1
“Tell me about a time you had a significant disagreement with a teammate over a technical direction. How did you resolve it?”
- 2
“Give an example of a project where you misjudged the complexity or timeline. How did you recover?”
- 3
“Suppose $X$ and $Y$ are mean zero, unit variance random variables. If the least squares regression of $Y$ against $X$ gives a slope of $\beta$, what is the slope of the regression of $X$ against $Y$?”
- 4
“If I break a stick of unit length into three random pieces, what is the expected length of the largest piece?”
- 5
“How would you detect overfitting in a machine learning model applied to noisy financial time series data?”
- 6
“What is the Delta of an at-the-money binary option as time approaches expiry?”
- 7
“Suppose the moon were to disintegrate and fall to earth over 5,000 years. How does this influx of power compare to that of the Sun?”
- 8
“If you are completely stuck on a live coding problem, what do you do?”
- 9
“What is a commonly accepted statistical assumption in finance that you believe is fundamentally flawed?”
- 10
“How would you price a used car market if you had no access to historical pricing data?”
- 11
“Describe a situation where your code or model failed in production. How did you handle it?”
- 12
“Give an example of a time you disagreed with a senior colleague or professor. How did you resolve the impasse?”
- 13
“How do you prioritize competing tasks when working across multiple technical projects?”
- 14
“Tell me about a time you noticed an error in someone else's work. How did you approach them?”
- 15
“Tell me about a time you failed to meet a goal. What did you learn?”
- 16
“How do you handle repetitive debugging tasks without losing attention to detail?”
- 17
“Tell me about a technical project where you underestimated the scope. How did you recover?”
- 18
“How do you maintain focus and accuracy when working on long, tedious analytical tasks?”
- 19
“Describe a time you identified a hidden bug that others missed. What was your methodology?”
- 20
“What is the probability that the final passenger on a 100-seat plane gets their assigned seat if the first passenger chose randomly?”
- 21
“If you roll a fair die repeatedly until you roll a 6, what is the expected number of rolls? What if you require two 6s in a row?”
- 22
“Consider all 100-digit numbers. What is the last digit of the sum of the products of their non-zero digits?”
- 23
“Explain the mathematical intuition behind Lasso ($L_1$) versus Ridge ($L_2$) regularization. Why does $L_1$ induce sparsity?”
- 24
“How do you evaluate a machine learning model designed for non-stationary time series financial data?”
- 25
“What is the Delta of an at-the-money binary option as time to maturity approaches zero?”
The rubric
How G-Research scores you
The interviewer is checking against a scorecard. Knowing the categories is half the battle.
Aggregation. Competencies scored: Analytical reasoning, coding precision, communication clarity, mathematical rigour, and intellectual humility. Aggregation: Interviewers submit independent written scorecards evaluating each competency on a structured scale. Hiring committees review these scorecards collectively.
How to practise
Live AI mocks tuned to the firm
Conversational mocks that read your CV, ask G-Research-specific questions, push back when answers are weak, and produce a scored report.
- Reads your CV first. Vyo pulls real lines from your CV ("walk me through that deal you mentioned") instead of asking the same five questions everyone gets.
- Follows up where answers are weak. Weak on numbers? Vyo digs. Vague on motivation? Vyo digs. No two runs are the same.
- Matches G-Research's interview style. Friendly chat, professional grilling, or full-pressure. Calibrated to how G-Research actually runs the round.
- Annotated transcript after every run. Strong lines green, wobbles amber, red flags red. Specific notes on what to fix and how to say it next time.
Lazard · IB Analyst
Technical Interview · CV-aware
Interviewer
I see from your CV you completed Spring Week at Lazard. Walk me through a deal you observed during that placement.Interviewer
You mentioned trading comps. Walk me through how the peer set was selected, what criteria were used, and were there any borderline cases?Full report when you end
Why candidates fail
How candidates lose the G-Research live round
Specific failure patterns documented across past applicants. Most are avoidable with focused preparation.
- 1
Inability to handle follow-up probes
Collapsing when an interviewer challenges an initial assumption instead of defending or adapting it logically.
- 2
Superficial CV knowledge
Listing advanced machine learning frameworks or algorithms on the CV that the candidate cannot explain from first principles.
- 3
Failure to structure thoughts
Writing code or solving math problems in complete silence without communicating intermediate hypotheses.
- 4
Poor calibration of scale
Providing unrealistic estimates during Fermi estimation or sizing questions.
- 5
Lack of firm-specific preparation
Treating G-Research as a generic software house rather than a systematic quantitative finance firm.
- 6
Defensiveness
Arguing with the interviewer when corrected on a mathematical error.
- 7
Inadequate coding mechanics
Writing syntactically invalid code or ignoring edge cases like null inputs and empty lists.
- 8
Generic closing questions
Asking lazy questions at the end that could be answered via a basic web search.
What works
What separates candidates who pass
Concrete moves drilled by candidates who clear the cut, drawn from applicant accounts and recruiter feedback.
Rigorous first-principles derivation
Starting from fundamental definitions rather than relying on memorised formulas.
Structured vocalisation
Explaining thought processes clearly and pausing to invite interviewer feedback.
Algorithmic efficiency awareness
Immediately discussing time and space complexity trade-offs for proposed code solutions.
Deep project familiarity
Knowing every line and assumption of their listed CV projects inside out.
Calm under pressure
Maintaining composure when presented with impossible-sounding brainteasers or stress tests.
Thoughtful inquiry
Asking probing questions about the firm's research infrastructure, computing scale, or data pipelines.
Intellectual honesty
Admitting gaps in knowledge immediately and pivoting to how they would investigate the problem.
Clean code delivery
Writing modular, readable, and bug-free code on the first pass during live technical tests.
What gets you through
Five moves that decide the interview
- 01Have a CV walkthrough rehearsed. Two-minute version of your CV that connects every role to why this firm. Most interviews open with "walk me through your CV". Knowing yours cold is the foundation.
- 02Three anchor stories. Prepare three behavioural stories that demonstrate multiple competencies each. Reuse them, reframe them. You will get further than candidates with one story per question.
- 03Plant follow-ups in your answers. End answers with a hook the interviewer can dig into. "Happy to walk through the modelling if useful" turns one question into a longer conversation on your terms.
- 04Reference G-Research concretely. Specific deal, division, recent news, a person you spoke to at an event. "I admire the brand" loses to "I followed your work on the X transaction".
- 05Have two smart questions ready. For the "any questions for me?" close. Not generic ("what is the culture like"), specific ("what is the typical analyst staffing model on a cross-border M&A deal here").
FAQ
G-Research interview questions, answered
How far in advance are live interviews scheduled?
Candidate reports suggest recruitment teams provide between 3 and 7 days' notice via email.
What technology stack is used for live coding?
Shared collaborative code editors embedded within video conferencing tools or specialized technical screening platforms.
What should I wear for video interviews?
Standard professional business attire is expected, matching the standard of an in-person corporate meeting.
Where should I focus my eye-line during video calls?
Look directly into the webcam lens when speaking to simulate direct eye contact with the interviewer.
What should I do if I experience technical difficulties during the call?
Notify the recruitment coordinator immediately via backup contact channels or email while attempting to rejoin.
Am I allowed to use scrap paper during technical problems?
Yes, keeping blank paper and a pen nearby for rough calculations and diagramming is strongly recommended.
What should I do if I am asked a question I do not know?
Acknowledge the gap honestly, state your initial assumptions, and explain how you would begin investigating the problem.
Are calculators permitted during quantitative interviews?
Mental arithmetic and algebraic simplification are expected; calculators are generally unnecessary and discouraged.
How long after the live interview should I expect feedback?
Candidate reports indicate response times range from several days to two weeks depending on panel availability.
Should I record the interview session?
No, recording company interviews without explicit written consent is strictly prohibited.
How technical are the questions for general software engineering tracks versus quantitative research?
Software tracks emphasize systems architecture, algorithms, and code optimization, whereas research tracks emphasize applied probability, statistics, and modelling.
Can I use IDE autocomplete features during live coding?
Standard minimalist shared code pads are typically used, meaning syntax must be written from memory without heavy IDE assistance.
The other rounds
The rest of the G-Research process
Live interview is one of four rounds. Practise each one free on Intervyo.
Practise free
Rehearse the G-Research live interview free
Practise every stage on Intervyo with AI-scored feedback: HireVue, psychometrics, live mock interviews, and CV. Free to start, no card required.
Start practising freeFree tools, upgrade any time
Intervyo is not affiliated with or endorsed by G-Research. Interview questions are sourced from past applicants and the firm's published guidance; verify on the firm's careers site. Sector: Graduate Scheme.
G-Research
Practise free