IBM interview questions

Showing 1–27 of 27 results

• Break Adjacent Duplicate CharactersMedium

  Tests string traversal, run-length counting, and greedy optimization to determine the minimum character replacements needed to eliminate adjacent identical characters across a list of words.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Calculate Revenue from Highest-Inventory VM SalesEasy

  Sell each VM unit at its current inventory count, always choose a type with maximum inventory, and compute total revenue safely.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Count Employees With Unbroken Renewable Access at the Last TimestampMedium

  A simulation problem about an access system where each employee's access lasts a fixed duration after every setting and can be extended only while it is still active. It asks how many employees hold valid access at the last timestamp, testing exact expiry boundaries, permanent lapses and per-employee state.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• Count Index Triplets Whose Max Minus Min Is at Most dMedium

  A counting problem that asks how many triplets of positions in an integer array have a spread, the largest value minus the smallest, of at most d. It tests handling duplicates and inclusive bounds, producing a count that can exceed 32 bits, and meeting time limits for arrays of up to 100,000 elements.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• Count Replacements Needed to Separate Equal Adjacent CharactersMedium

  This problem tests your ability to identify and process consecutive character runs in strings and apply greedy calculations to minimize replacement operations.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Count Seat-Price Trios Within a Budget SpreadMedium

  Count how many trios of resale-marketplace seat listings have prices within a fixed budget spread of each other, using sorting and a two-pointer sweep instead of an O(n^3) scan.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• Count Sensor Trios Within a Tolerance BandMedium

  Count how many groups of three readings stay inside a tolerance band (max minus min within a limit). Tests sorting, a sliding window, combinatorial counting and handling of duplicates and large totals.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• Count Tree Vertices with Pythagorean DistancesMedium

  Calculate distances from three fixed vertices to all other vertices in an unweighted tree and count those whose strictly positive distances form a Pythagorean triple.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Debug and Harden a Ticketing Backend APIMedium

  Implement and harden an in-memory ticket service backend to handle deterministic cursor pagination, idempotent reservations, cache invalidation, and oversell prevention.

  IBMSoftware EngineerSoftware Engineering FundamentalsOnline Assessment
• Describe open-source contribution experienceMedium

  This IBM software-engineer take-home behavioral question evaluates your open-source contribution experience end to end: the projects and repositories you worked on, your specific roles and most impactful contributions, how you collaborated with maintainers through issues, code reviews, and RFCs, and the lessons you took from code reviews and issue triage. A strong answer goes deep on one or two repositories with quantified, evidence-backed impact.

  IBMSoftware EngineerBehavioral & LeadershipOnline Assessment
• Design a lead-scoring modelEasy

  Design an end-to-end lead-scoring model for a marketing/growth data science team: define the label and scoring time, engineer leakage-safe features, compare a logistic-regression baseline with gradient-boosted trees, and evaluate with capacity-aware ranking metrics. The question also probes multicollinearity, model interpretability (SHAP/reason codes), routing-threshold design, and operational risks like selection bias, class imbalance, fairness, and drift.

  IBMData ScientistMachine LearningTechnical Screen
• Design a multithreaded CPU 1D convolutionHard

  An IBM software-engineer take-home that asks you to design and implement a multithreaded CPU version of valid 1D convolution across three regimes: a tiny kernel (K=3), a kernel as long as the input (K=N, a single dot product), and a 100-thread cap. It evaluates parallel decomposition, scheduling, cache locality, false-sharing avoidance, SIMD vectorization, and reduction/synchronization choices.

  IBMSoftware EngineerSystem DesignOnline Assessment
• Estimate free-trial conversion probabilityEasy

  This question evaluates a data scientist's competency in cohort definition, censored-data handling, segmentation and confounding assessment, and uncertainty quantification for measuring free-trial to paid conversion, and it belongs to the Analytics & Experimentation domain within data science.

  IBMData ScientistAnalytics & ExperimentationTechnical Screen
• Explain CI/CD, Kubernetes delivery, and troubleshootingEasy

  This question evaluates operational DevOps and cloud-native competencies, specifically CI/CD concepts and tools, GitOps delivery with Argo CD, Kubernetes deployment and troubleshooting (including Helm and StatefulSet vs Deployment), containerization rationale, Linux process and file inspection, network diagnostics, database deployment safety, and infrastructure concepts such as HPC and VPC. It is commonly asked in Software Engineering Fundamentals and DevOps interviews to measure production-readiness, troubleshooting ability, and tooling familiarity, emphasizing practical application while requiring underlying conceptual understanding of system behavior and reliability.

  IBMSoftware EngineerSoftware Engineering FundamentalsOnsite
• Find Customers Whose Average Website Resource Usage Exceeds 50 PercentMedium

  Practice PostgreSQL aggregation with HAVING, customer-level average resource utilization, strict thresholds, two-decimal rounding, and ascending email output.

  IBMSoftware EngineerData Manipulation (SQL/Python)Online Assessment
• Find minimum subarray length with k distinct integersMedium

  This question tests proficiency with the sliding window technique and hash map-based frequency tracking, applied to subarray problems requiring exact cardinality constraints. It assesses the ability to distinguish between "exactly k distinct" and "at most k distinct" variants — a nuance that reveals depth in algorithmic thinking commonly evaluated in software engineering interviews.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• Function Signature Lookup with Exact and Variadic Parameter MatchingMedium

  A coding question about an IDE-style signature lookup engine: given registered functions, some whose last parameter is variadic, return for each call the sorted names of every function that accepts its argument types. It tests exact type-sequence matching, zero-or-more variadic arguments and edge cases such as empty argument lists.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• How do you detect and fix multicollinearity?Easy

  This question evaluates understanding of multicollinearity and related competencies in regression diagnostics, feature engineering, model interpretability, and the trade-offs between coefficient inference and predictive performance.

  IBMData ScientistStatistics & MathTechnical Screen
• How would you use generative AI at work?Easy

  This question evaluates a candidate's judgment and leadership in applying generative AI tools, testing competencies in risk assessment, data confidentiality, security awareness, correctness verification, and team change management for a software engineer role.

  IBMSoftware EngineerBehavioral & LeadershipOnsite
• Manage Authentication Tokens with a Fixed Time to LiveMedium

  Design an authentication token manager that handles token generation, renewal, and active token counting based on a fixed time-to-live.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Maximize Revenue from Depleting VM InventoryMedium

  This problem tests your ability to efficiently manage and update a dynamic collection of items using a priority queue to always extract the maximum value.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Maximize Revenue from Depleting VM StockMedium

  Tests greedy optimization, sorting, and arithmetic progression math to compute maximum revenue from diminishing-value stock with large order quantities.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Maximize palindromic strings after character swapsMedium

  This question evaluates proficiency in string manipulation, combinatorial character-count reasoning, and redistribution of characters under constrained swap operations within the Coding & Algorithms domain.

  IBMSoftware EngineerCoding & AlgorithmsTechnical Screen
• Six Timed Behavioral Prompts: AI Tools, Debugging, Teamwork, Explaining, Goals, QualityMedium

  Six prompts from a recorded behavioral interview, each with one minute to prepare and three minutes to answer: vetting an AI tool's output, diagnosing a failure, teamwork, explaining your work, reaching a goal, and verifying that something you built works. Tests structured, specific storytelling under strict time limits.

  IBMSoftware EngineerBehavioral & LeadershipOnline Assessment
• Sum Pairwise Remainders Across Two Integer RangesEasy

  Calculate the total sum of remainders obtained by taking each integer in a dividend range modulo each non-zero divisor in a divisor range. The problem tests division properties, edge-case handling for zero divisors, and avoiding integer overflow.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Sum the XOR of Every Cross-Array PairEasy

  Tests bitwise frequency analysis and arithmetic optimization by reducing an O(N * M) cross-product XOR sum to an O(B * (N + M)) bit-by-bit contribution calculation.

  IBMSoftware EngineerCoding & AlgorithmsOnline Assessment
• Write SQL to compute max and min marksEasy

  This question evaluates a candidate's ability to perform basic data aggregation in SQL, specifically computing summary statistics from a single table. It is commonly asked to assess competence in data manipulation and query result shaping within the Data Manipulation (SQL/Python) domain, focusing on practical application of SQL aggregation rather than purely conceptual theory.

  IBMData ScientistData Manipulation (SQL/Python)Online Assessment