TredenceData Scientist

Updated Aug 23, 2026 · Reviewed by the Dataford team

Tredence Data Scientist interview questions & guide 2026

Every question Tredence interviewers actually ask, the frameworks that win the room, and the language hiring managers respond to.

6 rounds · ≈ 4-6 weeks

Talent Acquisition Call

Online Assessment

Automated Communication Test

Technical Interview Round 1

Technical Interview Round 2

HR/Leadership Evaluation

1. What is a Data Scientist at Tredence?

As a Data Scientist at Tredence, you occupy a critical position bridging sophisticated machine learning models, big data architecture, and strategic business consulting. Tredence specializes in last-mile AI adoption, delivering customized analytics solutions across verticals such as retail, CPG, supply chain, healthcare, and trust and safety. In this role, you do not build models in a vacuum; you design end-to-end data science solutions that solve high-stakes operational challenges—from demand forecasting and dynamic pricing optimization to real-time content moderation and buyer risk scoring.

The impact of this role is immediate and broad. You will work directly with enterprise client data and cross-functional teams comprising data engineers, product managers, and business stakeholders. Whether you are deploying Retrieval-Augmented Generation (RAG) pipelines on Azure Databricks, engineering statistical anomaly detection algorithms for high-volume marketplace transactions, or structuring causal inference studies to measure payment policy shifts, your work directly moves primary business metrics for Fortune 500 clients.

Joining Tredence offers a distinct technical environment characterized by variety, rigor, and rapid growth. Candidates who thrive here possess strong foundational knowledge in machine learning theory, fluently write production-grade Python and SQL, and exhibit sharp commercial instinct. You will be expected to defend your mathematical trade-offs, explain complex modeling concepts to non-technical executives, and build resilient pipelines that scale across millions of daily events.

2. Common Interview Questions

Interview questions at Tredence are drawn directly from real candidate experiences across online assessments, technical rounds, and managerial evaluations. While individual loops adapt to specific client verticals and seniority levels, the underlying patterns consistently emphasize foundational rigor, coding efficiency, and practical business analytics.

Product Sense & Case Studies

Questions in this category evaluate how effectively you translate broad, ambiguous business problems into structured analytical frameworks and end-to-end data science solutions.

Given a retail data science problem like demand forecasting or dynamic pricing, how do you structure the problem from data ingestion to production model deployment?
How would you design a seller risk framework on an e-commerce platform to detect fraudulent listings and reduce buyer chargebacks?

A enterprise client observes a sudden 15% drop in user conversion rate following a feature rollout. Walk me through your framework for metric drop diagnosis.
How do you handle trade-offs between model interpretability and predictive accuracy when presenting recommendations to client executives?
Walk through an end-to-end data science project from your resume, detailing the specific feature engineering choices, model selection criteria, and business outcomes achieved.
If a client provides incomplete or unlabelled transactional data under a tight deadline, how do you structure a high-value analytical response?

SQL & Data Manipulation

These questions test your ability to extract, transform, and manipulate complex, multi-table enterprise datasets using clean and performant code.

Given employee transactional data with monthly records, write a query using SQL window functions (such as SUM() OVER(PARTITION BY ...)) to compute cumulative salary totals over time.
Consider Table A with 10 rows and Table B with 20 rows matching on an ID column. What are the minimum and maximum row counts returned by an INNER JOIN, LEFT JOIN, and CROSS JOIN?
Write a Python script to check if a string containing nested brackets, such as [{}({})], is properly balanced.
How do you perform group-by aggregations and data scaling across a 100-feature dataset in pandas versus PySpark without running into memory bottlenecks?

A/B Testing & Experimentation

These questions measure your ability to design sound online experiments, choose correct testing metrics, and account for potential statistical anomalies.

How do you determine the required sample size and exposure duration for an A/B testing setup, and how do you verify statistical significance?
What are common experimentation pitfalls—such as novelty effects, network leakage, and Sample Ratio Mismatch (SRM)—and how do you mitigate them?
How would you design an experiment to evaluate a real-time AI content moderation model while ensuring minimal disruption to genuine buyer conversion rates?

Statistics & Probability

Evaluators assess whether you understand the underlying statistical mechanisms governing dataset distribution, modeling assumptions, and metric evaluation.

Explain the difference between $R^2$ and Adjusted $R^2$. What happens to both metrics when an irrelevant feature is added to a linear regression model?
Explain Type I and Type II errors in the context of loan default prediction. Which error type carries a higher commercial cost for a risk assessment model?
What is the Central Limit Theorem, and why is it fundamental when interpreting confidence intervals and p-values during hypothesis testing?

Product Metric Design & Metrics

These questions evaluate your capacity to formulate robust, actionable metrics that align algorithm performance with business goals.

How do you approach product metric design for a new Trust & Safety live-moderation engine? Define your leading and lagging performance indicators.
What evaluation metrics would you choose for a heavily imbalanced dataset (e.g., 95% non-defaulters vs. 5% defaulters), and why is accuracy a misleading metric here?

Machine Learning & Core Modeling

Technical screeners dig deeply into core machine learning algorithms to verify that you understand backend math and optimization techniques.

Explain the bias-variance tradeoff and contrast $L_1$ (Lasso) and $L_2$ (Ridge) regularization methods in terms of feature selection and penalty geometry.
Walk me through the mathematical mechanics of Convolutional Neural Network (CNN) architectures or the hyperparameter tuning steps for Random Forest and XGBoost.

Behavioral & Leadership

These questions focus on stakeholder management, communication clarity, domain ownership, and how you navigate project ambiguity or pushback.

Describe a situation where a client or interviewer aggressively challenged your modeling methodology or statistical findings. How did you handle the discussion?
Tell me about a time when you had to work with noisy, messy, or incomplete data under severe time constraints to deliver actionable business insights.
Give an example of how you explained a complex machine learning concept or statistical trade-off to a non-technical executive or client stakeholder.
Describe a project that failed to meet its target accuracy or deployment timeline. What root causes did you identify, and what did you learn from the experience?

Access the full Tredence Data Scientist prep plan

Every Data Scientist question, updated weekly
Model answers with SQL and Python solutions
Recent, real interview reports

Get my prep plan

03 · Question bank

The questions most likely to come up

Sorted by relevance to this company

Design Test for New FeatureMedium

Design an A/B test for a new platform feature, including success metrics, power, guardrails, and a clear ship decision.

experiment designfeature evaluationA/B Testing

Recently asked

Feature Engineering for Supervised ModelsEasy

Explain feature engineering and why transforming raw inputs can materially improve supervised model performance.

Cross-ValidationFeature EngineeringSupervised Learning

Recently asked

Access the full Tredence Data Scientist prep plan

Everything you need to walk in ready.

Get my prep plan

3. Getting Ready for Your Interviews

Preparing for a Data Scientist interview loop at Tredence requires a balanced focus on core computer science foundations, statistical theory, practical machine learning, and direct resume defense. The technical evaluations test both theoretical depth and production readiness, meaning surface-level familiarity with high-level libraries will not suffice.

To evaluate candidates consistently, Tredence interviewers rely on four core assessment criteria:

Role-Related Technical Competency – Evaluators assess your depth in fundamental computer science, Python scripting, complex SQL querying, and statistical modeling. You must demonstrate a clear understanding of machine learning algorithms from first principles rather than treating libraries as black boxes. Demonstrating proficiency in data manipulation and query efficiency is essential for passing initial technical screens.

Structured Problem-Solving & Business Acumen – Candidates are evaluated on their ability to translate vague business problems into well-defined data science workflows. Interviewers look for systematic case analysis, logical metric selection, explicit trade-off discussions, and an understanding of client context. Strong candidates explain why a particular model or feature transformation fits the specific business objective.

Resume & Project Ownership – A significant portion of the interview loop involves deep dives into your past projects. Interviewers will interrogate your architecture choices, feature selection methods, model evaluation metrics, and deployment pipelines. You must be prepared to articulate your individual contributions and defend your methodology under questioning.

Communication & Stakeholder Leadership – Working at an analytics consultancy requires clear communication across both technical and business audiences. Evaluators examine how concisely you explain complex algorithmic trade-offs, address constructive pushback, and structure executive presentations. Confidence, active listening, and technical clarity are vital attributes.

4. Interview Process Overview

The hiring process for a Data Scientist at Tredence is rigorous, multi-staged, and thoroughly technical. On average, candidates navigate 4 to 5 distinct evaluation stages over 2 to 4 weeks. The process is designed to evaluate raw analytical talent, hands-on coding proficiency, statistical intuition, and client-facing communication skills.

Initial screening typically begins with a talent acquisition call followed by a comprehensive Online Assessment (OA). The online test lasts approximately 90 to 120 minutes and covers a mix of multiple-choice questions (MCQs) spanning SQL, Python, probability, statistics, and machine learning theory, along with live coding challenges. Candidates who perform well on the technical assessment may also complete an automated AI-driven communication test evaluating spoken, written, and comprehension skills.

Upon passing the preliminary assessments, candidates advance to two separate live technical interview rounds. The first technical round focuses on live coding—demanding real-time SQL query writing and data structure manipulations—alongside deep dives into classical machine learning algorithms. The second technical round frequently takes a techno-managerial focus, concentrating on complex project walk-throughs, business case studies, metric evaluation, and system design. The loop culminates in an HR or leadership evaluation that checks cultural alignment, salary expectations, and operational fit.

06 · The loop

The interview process, end to end

≈ 4-6 weeks · 6 rounds

Talent Acquisition Call

Initial screening call to discuss the candidate's background and role fit.

Online Assessment

Comprehensive online test covering SQL, Python, probability, statistics, and machine learning theory.

Automated Communication Test

AI-driven test evaluating spoken, written, and comprehension skills for candidates who perform well in the online assessment.

Technical Interview Round 1

Live coding interview focusing on SQL query writing and classical machine learning algorithms.

Technical Interview Round 2

Techno-managerial interview concentrating on project walk-throughs, business case studies, and system design.

HR/Leadership Evaluation

Final evaluation checking cultural alignment, salary expectations, and operational fit.

The visual timeline above outlines the standard sequence of stages you will encounter in the Tredence interview funnel. Candidates should use this roadmap to structure their preparation, ensuring core coding skills and theoretical concepts are mastered prior to the online assessment, while reserving time for case study practice and resume deep dives ahead of the technical rounds.

5. Deep Dive into Evaluation Areas

To pass the technical panels at Tredence, you must demonstrate deep domain expertise across four primary technical evaluation areas.

SQL Data Manipulation & Querying Efficiency

SQL is evaluated heavily across both online assessments and live technical interviews. Interviewers test whether you can query enterprise schemas, write performant aggregations, and manipulate complex data structures without relying on temporary tables or inefficient subqueries.

Be ready to go over:

SQL Window Functions – Mastery of ROW_NUMBER(), RANK(), DENSE_RANK(), LEAD(), LAG(), and aggregate window functions using OVER(PARTITION BY ... ORDER BY ...) clauses.

Complex Joins and Null Handling – Predicting row counts across INNER, LEFT, RIGHT, and FULL OUTER joins, understanding NULL behavior in equality checks, and handling duplicate keys.
Conditional Aggregations & Pivoting – Utilizing conditional CASE WHEN statements inside aggregate functions to pivot long-format data into wide business summaries.
Advanced concepts (less common) – Recursive Common Table Expressions (CTEs), query plan optimization, indexing strategies, and window frame specifications (ROWS BETWEEN).

Example scenarios:

"Write a query to calculate the 3-month rolling average salary for each department, partition by department ID, and handle missing months gracefully."
"Given two tables with duplicate IDs, calculate the exact row count produced by a LEFT JOIN and explain how WHERE clause placement changes the output."

Product Sense, Metric Design & Drop Diagnosis

Candidates must prove that they can connect machine learning outputs to operational revenue, user retention, and risk management. You will be presented with open-ended business scenarios where you must structure the problem, design tracking metrics, and diagnose sudden operational anomalies.

Be ready to go over:

Product Metric Design – Establishing clear primary, secondary, and guardrail metrics for new features or Trust & Safety systems (e.g., balancing conversion rate against chargeback rates).
Metric Drop Diagnosis – Methodically isolating root causes when a core key performance indicator (KPI) drops unexpected (e.g., segmenting by geography, device, API version, user cohort, or pipeline latency).
Data Pipeline Anomaly Detection – Identifying whether a sudden metric shift stems from genuine user behavior change or underlying data pipeline and logging failures.
Advanced concepts (less common) – Designing counterfactual evaluations, establishing leading indicator lead-times, and constructing multi-metric composite health scores.

Example scenarios:

"A retail client experiences a sudden 12% drop in online checkout conversions immediately after deploying a new recommendation model. Walk me through your step-by-step diagnostic workflow."
"Design a scorecard of leading metrics to monitor seller risk and item quality on a live marketplace without introducing friction for legitimate high-volume sellers."

Experimentation, A/B Testing & Causal Inference

Understanding how to validate data science hypotheses through rigorous offline and online testing is fundamental at Tredence. You must demonstrate expertise in experiment setup, statistical evaluation, and detecting potential testing failures.

Be ready to go over:

A/B Testing Methodology – Formulating null and alternative hypotheses, selecting primary evaluation metrics, calculating minimum detectable effect (MDE), and sizing sample populations.
Statistical Significance & Hypothesis Testing – Choosing appropriate test statistics ($z$-test, $t$-test, Chi-Square), interpreting p-values, and establishing confidence boundaries.
Experimentation Pitfalls – Identifying and resolving Sample Ratio Mismatch (SRM), novelty or primacy effects, network spillover effects, and multiple testing correction ($p$-hacking).
Advanced concepts (less common) – Difference-in-Differences (Diff-in-Diff), Synthetic Control methods, Causal Trees, and variance reduction techniques like CUPED.

Example scenarios:

"You run an A/B test on a dynamic pricing algorithm. The test variant shows a statistically significant increase in revenue but a drop in total order volume. How do you decide whether to launch?"
"Explain how you would detect and resolve a Sample Ratio Mismatch (SRM) during an ongoing experiment."

Applied Machine Learning & Predictive Modeling

Interviewers conduct exhaustive technical evaluations on statistical modeling and classical machine learning algorithms. You must be capable of explaining backend mechanics, hyperparameter optimization, and evaluation metric selection.

Be ready to go over:

Core Supervised Learning – Linear and Logistic Regression assumptions, Decision Trees, Ensemble methods (Random Forest, XGBoost, LightGBM), and hyperparameter tuning strategy.
Model Evaluation & Imbalanced Data – Choosing between Precision, Recall, $F_1$-Score, ROC-AUC, and PR-AUC, and handling severe class imbalances through stratification or resampling.
Unsupervised Learning & Feature Reduction – $K$-Means clustering, Silhouette score interpretation, Elbow method mechanics, and Principal Component Analysis (PCA).
Advanced concepts (less common) – Large Language Model (LLM) fine-tuning, Retrieval-Augmented Generation (RAG) pipelines, Azure Databricks workflow orchestration, and ML explainability (SHAP / LIME).

Example scenarios:

"Why is $R^2$ alone insufficient to assess a linear regression model? How does Adjusted $R^2$ penalize feature addition, and what assumptions must be verified?"
"Explain how you would build, evaluate, and monitor a time-series forecasting model for inventory planning across thousands of retail store locations."

Access the full Tredence Data Scientist prep plan

Every Data Scientist question, updated weekly
Model answers with SQL and Python solutions
Recent, real interview reports

Get my prep plan

08 · Topic breakdown

What they actually test for

Topic distribution

All topics

Machine Learning (ML) FundamentalsSQL (Core)Deep Learning (DL) FundamentalsNatural Language Processing (NLP)Model Evaluation & Performance Assessment

6. Key Responsibilities

As a Data Scientist at Tredence, your daily work centers on delivering end-to-end analytics and artificial intelligence solutions that address complex client problems. You will rarely work in isolation; instead, you operate as part of a high-performing pod that integrates closely with enterprise client stakeholders, engineering leads, and business analysts.

Your core deliverables span the entire data science project lifecycle. In the initial project phase, you collaborate with business partners to translate ambiguous operational goals into formal analytical definitions. You write complex SQL scripts and Python pipelines to clean, aggregate, and explore massive, multi-terabyte datasets housed in environments like Azure Databricks, Snowflake, or AWS.

During the modeling phase, you select, train, and validate statistical and machine learning models—ranging from classical time-series forecasting (ARIMA, Prophet) and tree-based ensembles (XGBoost) to modern NLP and GenAI architectures (RAG, custom embeddings). You are responsible for rigorous model evaluation, conducting error analysis, checking feature importance, and ensuring models operate without unintended bias or overfitting.

Finally, you work with MLOps and data engineering teams to transition validated models into production environments. You help establish model monitoring frameworks to track prediction drift, latency, and cost efficiency. Throughout this process, you create executive dashboards, present findings to senior client leadership, and document reproducible analytical standards for the team.

7. Role Requirements & Qualifications

Candidates who successfully secure offers at Tredence bring a strong combination of formal quantitative training, practical programming skills, and business domain awareness.

Must-have technical skills – Advanced proficiency in Python (pandas, NumPy, scikit-learn, statsmodels) and SQL (window functions, CTEs, query optimization). Solid foundational knowledge of classical machine learning algorithms (linear/logistic regression, Random Forest, Gradient Boosting) and statistical concepts (hypothesis testing, distribution analysis, p-values).
Nice-to-have technical skills – Experience with cloud platforms (Azure, AWS, GCP), Azure Databricks, PySpark, MLflow, data visualization tools (Tableau, Power BI), deep learning frameworks (PyTorch, TensorFlow), and recent exposure to GenAI frameworks (LangChain, vector databases, RAG pipelines).
Experience level – Typically a Bachelor’s or Master’s degree in Computer Science, Data Science, Statistics, Mathematics, Operations Research, or an equivalent quantitative discipline. Roles range from mid-level Data Scientist (2–5 years of experience) to Senior Data Scientist (5–8+ years) and Data Science Manager.
Soft skills & stakeholder management – Exceptional structured problem-solving skills, ability to manage client expectations, strong technical writing capabilities, and a collaborative approach when working alongside cross-functional engineering teams.

Tip

While deep technical skills in machine learning are vital, Tredence heavily weights client-readiness and business communication. Demonstrating how your models directly impact revenue or operational efficiency is key to clearing senior tech-managerial rounds.

8. Frequently Asked Questions

Q: How technical is the Tredence Data Scientist interview compared to traditional tech companies?
A: The loop is equally rigorous on core statistical math, SQL, and Python coding, but places a much stronger emphasis on practical business case studies, metric interpretability, and past project defense rather than purely abstract LeetCode algorithms.

Q: What is the single most critical area to prepare for technical interviews?
A: Your past projects detailed on your resume. Tredence interviewers perform exhaustive resume deep dives, probing your exact choice of algorithms, baseline metrics, feature engineering techniques, and business impact.

Q: Are questions about Generative AI and LLMs mandatory for all Data Scientist roles?
A: While core machine learning, statistics, and SQL form the core of every loop, recent interviews increasingly incorporate questions on NLP, embeddings, RAG architectures, and LLM evaluation, particularly for senior roles working on cloud platforms like Azure.

Q: How quickly do candidates receive feedback between interview rounds?
A: Feedback turnaround is generally swift, often provided within 24 to 48 hours after each round. However, total loop duration can range from 2 to 4 weeks depending on client project alignment and scheduling availability.

Q: Does Tredence allow remote or hybrid work setups for Data Scientists?
A: Work arrangements depend on the assigned account and regional office location (e.g., Bengaluru, Hyderabad, Gurgaon, or US client sites). Most delivery teams operate under structured hybrid models balancing remote work with client collaboration.

9. Other General Tips

Master SQL window functions thoroughly: Expect live coding or online assessment scenarios requiring ROW_NUMBER(), DENSE_RANK(), LEAD(), LAG(), and conditional aggregations. Writing clean, efficient queries on the first pass creates a strong impression.
Structure your case study responses using structured frameworks: When presented with open-ended business scenarios, state your assumptions, define primary and guardrail metrics, outline data requirements, and discuss model trade-offs systematically before jumping into algorithm details.
Be ready to explain ML concepts to a layperson: Practice explaining complex topics—such as backpropagation, $L_1$/$L_2$ regularization, or ROC-AUC curves—using simple analogies. Interviewers test whether you can communicate findings to non-technical client leads.

Note

Avoid listing technologies or algorithms on your resume that you cannot defend in detail. Interviewers routinely pick specific tools from your resume and ask backend implementation questions.

Brush up on classical statistics, not just machine learning: Be prepared to write out formulas and explain the underlying assumptions of Linear Regression, p-value calculations, $R^2$ mechanics, and Type I / Type II errors clearly.
Clarify problem constraints before coding: During live coding or system design portions, ask clarifying questions regarding data distributions, missing values, memory limits, and target latency before writing code.

10. Summary & Next Steps

Targeting a Data Scientist role at Tredence offers an incredible opportunity to solve high-impact analytics challenges for leading enterprise organizations worldwide. The interview loop is designed to identify well-rounded technical professionals who pair quantitative rigor in SQL, Python, and statistical modeling with strong commercial instinct and client leadership.

To maximize your performance, focus your preparation on core computer science foundations, performant data manipulation, hands-on ML problem-solving, and crisp articulation of your past project choices. Rehearse step-by-step frameworks for metric drop diagnosis, experiment design, and structured case studies so that you can navigate open-ended technical questions with confidence.

You can explore additional interview insights, detailed practice questions, and candidate preparation resources on Dataford to sharpen your preparation for Tredence and similar analytics loops.

14 · Compensation

What this role pays

6 reports

USUSD

Estimated total compLow confidence · 6 data points

$0k-$0k

Median $148k / year

Base salary · 100%Stock (RSU) · 0%Cash bonus · 0%

25thEntry / smaller markets

$125k

50thTypical offer

$148k

90thTop performers / major metros

$170k

Breakdown by component

Base salary

100% of total

$125k$170k

$148k

median

Stock (RSU)

0% of total

$0$0

median

Cash bonus

0% of total

$0$0

median

Aggregated from 6 self-reported salaries via Glassdoor. Estimates only. Verify against your offer.

The compensation module above illustrates typical base and overall packages for Data Scientist positions at Tredence. Total compensation scales with seniority, domain specialization (such as GenAI or supply chain optimization), and geographical location. Candidates should evaluate these figures against their experience level when navigating offer discussions.

15 · More at this company

Other roles at Tredence

Product Manager Solutions Architect Account Executive GenAI Engineer Data Engineer Business Analyst

17 · FAQ

Tredence Data Scientist interview FAQ

Answered from real candidate and compensation data

How hard are Tredence Data Scientist interviews, and what offer rate should candidates expect?

In candidate-reported experience, Tredence Data Scientist interviews are most commonly rated as average difficulty, with 60 reported interviews. The reported offer rate is 33%, so competition exists, but it is not described as uniformly hard across all candidates.

What are the interview rounds for Tredence Data Scientist, and what happens in each stage?

The process starts with a Talent Acquisition Call to discuss background and role fit, followed by an Online Assessment. If you do well in the online assessment, you can get an Automated Communication Test, then two Technical Interview rounds. The loop ends with an HR/Leadership Evaluation that checks cultural alignment, salary expectations, and operational fit.

What does the Tredence Data Scientist online assessment test?

The Online Assessment is a comprehensive test that covers SQL, Python, probability, statistics, and machine learning theory. Preparation should therefore include both coding fundamentals and core ML and stats concepts, not just one area.

What technical topics are most important for Tredence Data Scientist interviews?

The most frequently emphasized topics include ML Fundamentals, SQL (Core), Deep Learning Fundamentals, NLP, and model evaluation and performance assessment. You should also be ready for SQL querying and data manipulation plus Python coding for DS, including building and training ML models through project work.

What is the Tredence Data Scientist pay range, and does it vary by level and location?

Compensation reporting shows a base minimum of $125k and a total maximum of $170k for Data Scientist roles. Pay varies by level and location, so candidates should expect different offer structures rather than one fixed number.

How should I prepare for the live technical interviews at Tredence for Data Scientist?

Technical Interview Round 1 focuses on live coding with SQL query writing and classical machine learning algorithms. Technical Interview Round 2 is more techno-managerial, emphasizing project walk-throughs, business case studies, and system design.

Tredence Data Scientist interview questions & guide 2026

1. What is a Data Scientist at Tredence?

2. Common Interview Questions

Product Sense & Case Studies

Access the full Tredence Data Scientist prep plan

The questions most likely to come up

3. Getting Ready for Your Interviews

4. Interview Process Overview

The interview process, end to end

5. Deep Dive into Evaluation Areas

SQL Data Manipulation & Querying Efficiency

Access the full Tredence Data Scientist prep plan

What they actually test for

6. Key Responsibilities

7. Role Requirements & Qualifications

Tip

8. Frequently Asked Questions

9. Other General Tips

Note

10. Summary & Next Steps

What this role pays

Other roles at Tredence

Other Data Scientist guides

Tredence Data Scientist interview FAQ