IIT Kanpur AI research interviews reported by candidates center on fundamentals, past research, and fit with a lab or panel. Some include a written test, and one describes a paper-reading discussion.
$7K
Avg. Base Comp
$9K
Avg. Total Comp
Not reported
Typical Rounds
Not reported
Process Length
Candidates describing IIT Kanpur AI research interviews report conversations that test how you reason about fundamentals and your own research, not LeetCode-style coding. The reports cover different titles, including research intern, researcher, research assistant and PhD applicant, so expect the format to vary by lab and role.
Several candidates mention a written test before the interview. One researcher candidate describes a test on mathematical aptitude and subject knowledge, and a PhD applicant describes GATE-level questions. The interviews that follow are described as friendly but probing. One researcher candidate faced a panel of six for 45 to 60 minutes, with questions on linear algebra, two-phase locking, concurrency and operating systems. A PhD candidate was asked for a formal definition of random variables and to solve small numerical problems on the board.
Research discussion features in nearly every report. Candidates were asked about past projects, research interests, and, for the researcher role, a proposed research plan and the resources it would need. A research intern describes being handed a paper mid-interview and asked to build intuition about it, with CNN architectures and bias in image classification as follow-ups. Another intern describes a casual call with a professor about motivation, time commitment and the experiments they could run in the lab.
Prepare by reviewing math and core CS basics, rehearsing a concise account of your projects, and practicing reading an unfamiliar paper aloud. One PhD candidate advises being upfront about gaps in your knowledge rather than bluffing. Candidates report both offers and rejections, and one rejected candidate found the written-test scoring opaque.
Synthesized from 6 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Indian Institute Of Technology, Kanpur process.
I walked in expecting to meet a professor, but my interviewer turned out to be a PhD assistant assigned to evaluate my AI and ML fundamentals. That caught me a bit off guard, but it actually made the conversation feel less formal and more collaborative. She started by asking about my background and then moved into my projects—nothing tricky there, just getting a sense of what I'd actually worked on. The interesting part came when she handed me a research paper partway through and asked me to get some intuitive understanding of it on the spot. I had to think through the concepts without memorizing jargon, which felt more like a genuine research discussion than a quiz. In a second round, she came back to that paper and dug deeper, asking what I'd understood and how I'd interpret certain ideas. The technical questions were focused on CNN architectures and image classification—specifically whether I'd worked with them and what I knew about bias in that context. Nothing algorithmic or LeetCode-style; it was all about whether I could think about practical research problems and understand the papers I claimed to care about. The whole thing lasted longer than a typical screening but shorter than a full technical interview. By the end, it felt less like an interrogation and more like a genuine conversation with someone who wanted to know if I could think like a researcher.
Prep tip from this candidate
Be ready to read and discuss a research paper in real-time during the interview—they're genuinely interested in how you think through concepts, not whether you memorize terminology. Have concrete examples from your own projects ready, and prepare to articulate what you actually understand about bias and limitations in AI systems, especially in domains like image classification.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Indian Institute Of Technology, Kanpur
Explain what a p-value is to someone who is not technical
| Question | |
|---|---|
| Hurdles In Data Projects | |
| Using R Squared | |
| Encoding Categorical Features | |
| Bias - Variance Tradeoff and Class Imbalance in Finance | |
| Slow SQL Query | |
| Model Product Performance Degradation | |
| CNNs vs Intensity-Based Features | |
| Assumptions of Linear Regression | |
| Coefficients of Logistic Regression | |
| Last Element of a Singly Linked List | |
| Classification and Regression | |
| Training vs Validation vs Test Data | |
| Data Preparation for Imbalanced Data | |
| Vision Setting and Execution Strategy | |
| Multicollinearity in Regression | |
| Your Strengths and Weaknesses | |
| Stakeholder Communication | |
| Testing Constraints | |
| Data Cleaning Experiences | |
| Why Do You Want to Work With Us | |
| Evaluate News | |
| Credit Score Estimation | |
| Score Based on Review | |
| 2nd Highest Salary | |
| Merge Sorted Lists | |
| Experiment Validity | |
| Employee Salaries (ETL Error) | |
| Decreasing Comments | |
| Scrambled Tickets |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report different entry routes. One intern emailed professors whose work matched their interests and heard back from one after about a month. Another candidate applied through the IIT Kanpur website and waited for a call letter. Routes may differ by role and lab.
Several candidates report a written test before the interview. One described it as mathematical aptitude and subject knowledge, another as GATE-level fundamentals, and a third as balanced but with unclear scoring. Not every candidate mentions a written test, so it may depend on the role.
Candidates report conversational but probing interviews with professors, research scholars or a panel. Topics reported include linear algebra, two-phase locking, concurrency, operating systems, random variables, numerical problems on a board, and CNN-based image classification. Interviewers reportedly care about reasoning more than recall.
Candidates typically describe being asked about previous research, interests and fit. A researcher candidate discussed a proposed research plan and needed resources. An intern was asked about time commitment and lab experiments, and another intern discussed a research paper handed over during the interview and revisited in a second round.
