149 questions from 16 interviews · updated from reports 2014-2024
About
Target Corporation is a US retail company that runs Target stores and online shopping, selling groceries, clothing, home goods, electronics, and other everyday products. In India, it is known for hiring software engineers, data analysts, and interns for technology and business teams.
The roles that come up most are Software Engineer, Data Analyst and Intern. This covers 16 candidate interviews reported from 2014 to 2024. Most sat it at entry level (14 of 16 that recorded a level), with 1 internship interviews alongside. Among the 11 that recorded either route, arrivals split between campus drives (11, 100%). Most questions fall under CS fundamentals and DSA.
Ans. Model it as a maximum flow problem from the oil source to the destination. Store the pipeline network as an adjacency list with residual capacities, then run Dinic’s algorithm using BFS levels and DFS blocking flows. The answer is the total flow pushed. Time complexity is O(EV²) in the standard bound.
Ans. 1NF means each column holds atomic values, 2NF means every non-key attribute depends on the whole primary key, and 3NF means non-key attributes do not depend on other non-key attributes. The key point is that each step removes redundancy and update anomalies by ensuring facts are stored in the right table.
Ans. Load balancing distributes incoming requests across multiple servers so no single server is overloaded and the service stays available. A load balancer sits in front of the servers and chooses a target using rules such as round robin, least connections, or health checks. The key detail is that unhealthy servers are removed from rotation.
Ans. Deletion in a heap is usually done by removing the root, replacing it with the last element, reducing the heap size, and heapifying down to restore the heap property. For deleting an arbitrary element, replace it with the last element, then heapify up or down as needed. The time complexity is O(log n).
Ans. OOPS in Python means modelling programs using classes and objects, with encapsulation, abstraction, inheritance and polymorphism. A class defines data and behaviour, while an object is an instance. Encapsulation groups state with methods, inheritance reuses or extends behaviour, and polymorphism lets different objects respond to the same method call appropriately.
Ans. Use the SQL window function PERCENT_RANK over an ordered set, usually with an OVER clause and ORDER BY. It calculates relative standing as rank minus one divided by total rows minus one within the partition. The key detail is that ties share the same rank, so gaps can appear.
Ans. Servers communicate by sending data over a network using agreed protocols, usually on top of TCP/IP. One server opens a connection to another server’s IP address and port, sends a request, and receives a response. Protocols such as HTTP, gRPC, or message queues define the data format, routing, and reliability expectations.
Ans. Function overriding is when a subclass provides its own implementation of a method already defined in its superclass, using the same method signature. Its main advantage is runtime polymorphism: code can call the superclass type while the actual subclass behaviour runs, making programs easier to extend, reuse, and maintain.
Ans. If each node has a parent pointer, the grandparent is simply node.parent.parent, provided both exist. If not, traverse from the root using DFS or BFS while carrying the current node’s parent and grandparent. When the target is found, return the stored grandparent. This takes O(n) time and O(h) space with DFS.
Ans. The logical order is FROM and JOIN, WHERE, GROUP BY, HAVING, SELECT, DISTINCT, ORDER BY, then LIMIT or OFFSET. The most important detail is that this is logical processing order, not necessarily the physical execution plan, because the database optimiser may reorder operations while preserving the same result.
Ans. Add the relevant database driver and Spring Boot starter, then configure the datasource URL, username, password and driver in application.properties or application.yml. Spring Boot auto-configures a DataSource from these settings. For JPA, add spring-boot-starter-data-jpa, define entities and repositories, and inject repositories or JdbcTemplate into services.
Ans. A HashMap is a key value data structure that uses hashing to store and find values quickly. Internally, it hashes the key to choose a bucket in an array. If multiple keys map to the same bucket, it resolves collisions, commonly with a linked list or tree, using equality checks to find the right entry.
Ans. Inheritance in Java lets one class extend another, reusing and specialising its fields and methods. An abstract class is a class that cannot be instantiated and may contain abstract methods that subclasses must implement. The key point is that Java supports single class inheritance, so a class can extend only one class.
Ans. A circuit breaker is a resilience pattern that stops calls to a failing service to avoid cascading failures. Its states are closed, where calls pass normally; open, where calls fail fast; and half-open, where limited trial calls are allowed to check recovery. Failures trip it open, and successful probes close it again.
Ans. Select the distinct salaries, order them in descending order, then skip N minus 1 rows and return the next one. The key detail is using distinct values, otherwise duplicate salaries can make the result wrong. In databases with window functions, DENSE_RANK over salary descending is the usual robust approach.
Ans. SQL queries are used to read, insert, update and delete data in relational tables. SELECT chooses columns, FROM chooses tables, WHERE filters rows, JOIN combines tables, GROUP BY groups results, HAVING filters groups, ORDER BY sorts results, and LIMIT restricts output. The key point is that clauses define each step of the result.
Ans. Paging is a memory management technique, a semaphore is a synchronisation primitive, and deadlock is a state where processes wait forever. Paging maps virtual memory to fixed-size physical frames. Semaphores control access to shared resources using counters. Deadlock needs mutual exclusion, hold and wait, no pre-emption, and circular wait.
Ans. Use a postorder DFS and compute the height of each subtree while updating a global maximum leaf-to-leaf distance. For each node with both children, the best path through it is left height plus right height, measured in edges. Return one plus the larger child height. Time is O(n), space is O(h).
Ans. The Diamond Problem is the ambiguity caused when a class inherits the same method through two different parent paths. Java avoids it for classes by not allowing multiple class inheritance. With interfaces, default methods can create a similar conflict, and Java requires the implementing class to override the method and choose the behaviour explicitly.
Ans. Choose a real conflict with meaningful stakes, such as priorities, ownership, or technical direction. Emphasise listening first, separating facts from assumptions, aligning on project goals, and agreeing clear actions. Show you stayed professional and followed up. Interviewers listen for maturity, collaboration, accountability, and evidence that the conflict improved the outcome.
Ans. A Singleton class in Java allows exactly one object of that class to exist and provides a global access point to it. Implement it with a private constructor, a private static final instance, and a public static getInstance method. The stored instance is the only data needed, and access time is O(1).
Ans. Use the shortest string as the base and binary search the answer length, checking whether any substring of that length appears in every other string. Store rolling hashes of substrings in hash sets, intersect candidates across strings, then print a verified candidate. Time is about O(total characters log minLength), with hashing.
Ans. Common sorting algorithms include bubble, selection and insertion sort at O(n²), merge sort and heap sort at O(n log n), and quicksort at average O(n log n) but worst O(n²). Merge sort needs extra space, heap sort is in place, quicksort is usually fastest in practice, and counting/radix sort can be linear for limited integer ranges.
Ans. Merge Sort is a divide and conquer sorting algorithm that splits the array into halves, recursively sorts each half, then merges the sorted halves. The key detail is that merging is linear in the number of elements. Its time complexity is O(n log n) in best, average, and worst cases, with O(n) extra space.
Ans. You can pivot a table using conditional aggregation: group by the row key and compute each output column with an aggregate over a CASE expression. For example, each target column sums or counts values only when a category matches. The key detail is that the pivoted column values must be known, unless you build dynamic SQL.
Ans. Concatenate the two arrays into one array, then sort the combined array. This is usually the simplest and most practical approach for unsorted input. If the arrays have lengths m and n, the time complexity is O((m + n) log(m + n)), with space depending on the sorting algorithm used.
Ans. I would build search around an inverted index, using Elasticsearch, OpenSearch, or a similar engine fed from the main database via events or batch sync. The key detail is keeping indexing separate from serving traffic, with tokenisation, stemming, ranking, filters, pagination, caching for common queries, and monitoring for freshness and relevance.
Ans. Fragmentation is wasted memory caused when free or allocated space is split or unused inefficiently. Internal fragmentation occurs when an allocated block is larger than requested, leaving unused space inside it. External fragmentation occurs when free memory exists in small scattered blocks. It happens due to variable-sized allocation, deallocation, and long-running memory reuse patterns.
Ans. Object-oriented programming models software as objects that combine data and behaviour. For example, a Car object has data like speed and fuel, and methods like brake or accelerate. Encapsulation protects data, inheritance lets ElectricCar reuse Car features, polymorphism lets different vehicles implement start differently, and abstraction hides engine details from the driver.
Ans. Scan the prefix expression from right to left using a stack. When you see an operand, push it. When you see an operator, pop the top two strings, concatenate them as first operand, second operand, then operator, and push back. At the end, the stack top is postfix. Time and space are O(n).
Ans. The Collection interface in Java is the root interface for groups of objects, defining common operations like add, remove, size and iteration. Its main subinterfaces are List for ordered, indexed elements, Set for unique elements, Queue for processing elements in order, and Deque for double-ended queues. Map is related but not a Collection.
Ans. Parse the JSON with Python’s json module into dictionaries and lists, extract the needed fields by key or iteration, then pass the formatted values to a PDF library such as ReportLab or FPDF. The key detail is to validate missing or unexpected fields before rendering. Processing is usually linear in the JSON size.
Ans. Use Floyd’s cycle detection algorithm with two pointers, slow and fast, to detect a loop in a linked list. Move slow one step and fast two steps each time. If they meet, a loop exists. If fast reaches null, there is no loop. It uses constant extra space and runs in linear time.
Ans. Replace every repeating group or multi-valued column with separate rows or a separate child table, so each field contains one atomic value. Keep the original table’s key in the child table as a foreign key, and use it with the extracted value, or a new identifier, to preserve uniqueness and relationships.
Ans. I would apply machine learning to personalise product recommendations, improve search ranking, predict demand, detect fraud, and optimise marketing. The most important detail is the feedback loop: collect clean behavioural and transaction data, train models offline, serve predictions with low latency, and validate impact through A/B tests on conversion, revenue, and customer retention.
Ans. Pick a real team situation with a clear goal, shared pressure, and a disagreement you helped resolve. Emphasise how you listened, separated facts from opinions, stayed calm, and moved the group towards a decision. Interviewers listen for self-awareness, respect for others, accountability, and evidence that the outcome improved because of your approach.
Ans. Select the distinct salaries, sort them in descending order, and return the second row using an offset. The key detail is using distinct, so duplicate top salaries do not hide the true second highest salary. This approach sorts the salary values, so its typical time complexity is O(n log n).
Ans. Google ranks pages mainly by relevance to the query and the authority or quality of each page. The classic idea is PageRank: a page is more important if many important pages link to it. Modern ranking also uses content, freshness, location, usability, spam detection and many other signals.
Ans. I would profile the column, trim obvious whitespace, convert valid strings to a proper DATE or TIMESTAMP type using safe parsing, and quarantine rows that fail conversion. The key detail is never overwriting raw values until invalid, ambiguous, null, and out-of-range dates have been reviewed or handled by clear rules.
Ans. Use Customers and Orders joined on customer_id: INNER JOIN returns only customers with matching orders, LEFT JOIN returns all customers and matching orders, RIGHT JOIN returns all orders and matching customers, and FULL OUTER JOIN returns every row from both sides. The key detail is unmatched columns become NULL in outer joins.
Ans. Yes, Python supports multithreading, but in CPython the Global Interpreter Lock allows only one thread to execute Python bytecode at a time. This limits CPU-bound parallelism, but threads still help with I/O-bound work. A mutex lock protects one shared resource, while a semaphore allows a fixed number of concurrent accesses.
Ans. Use a min heap of size n while scanning the array once. Add each number until the heap has n items, then replace the smallest heap item when a larger number appears. At the end, the heap contains the n highest numbers. Time complexity is O(m log n), space is O(n).
Ans. Choose a real example where your idea faced reasoned disagreement, not personal conflict. Emphasise listening first, clarifying the discrepancy with data or user impact, inviting critique, and adapting your proposal. Interviewers listen for collaboration, evidence-based persuasion, respect for teammates, and willingness to change your view rather than simply “win” the argument.
Ans. Arrays are generally better for searching, while linked lists are better for insertion when the insertion position is already known. Arrays allow direct indexing and, if sorted, binary search in O(log n). Linked lists need O(n) search, but can insert in O(1) by changing pointers, unlike arrays which may need shifting.
Ans. A database is designed for day-to-day transactions, while a data warehouse is designed for analysis and reporting over large historical data. Databases are usually normalised and frequently updated. Data warehouses combine data from multiple sources, store cleaned historical data, and help businesses run complex queries, spot trends, and make decisions without slowing operational systems.
Ans. A LEFT OUTER JOIN returns every row from the left table, matching rows from the right table, and NULLs for right-side columns where no match exists. INNER JOIN returns only matches, RIGHT JOIN keeps all right rows, FULL OUTER JOIN keeps all rows from both sides, and CROSS JOIN returns every combination.
Ans. A primary key uniquely identifies a row, a foreign key links one table to another, and a surrogate key is an artificial primary key with no business meaning. In an orders system, CustomerID may identify customers, Order.CustomerID is a foreign key, and an auto-generated OrderID is a surrogate key used because it is stable and simple.
Ans. The query reaches MySQL because the MySQL server process is already running, usually as a background service or daemon started by the operating system. NetBeans is only a client. It uses a JDBC driver to open a connection, usually over TCP to port 3306 or a local socket, and sends the SQL to mysqld.
Ans. Use a self-balancing binary search tree, such as an AVL tree, Red Black tree, Treap, or skip list. Store smaller elements on the left and larger on the right, rebalance after updates, and return elements by inorder traversal. Insert, delete, and search take O(log n), traversal takes O(n).
Ans. Use realloc to increase an already allocated memory block, by passing the existing pointer and the new required size. It may extend the same block or allocate a new one and copy existing data. The maximum size is not fixed by C. It is limited by available heap, address space, and SIZE_MAX.
Ans. Pick a real example where you had invested effort but stayed open to feedback. Emphasise that you asked why the solution was rejected, separated ego from the decision, compared options against team goals, and adapted quickly. Interviewers listen for humility, collaboration, resilience, sound judgement, and learning rather than defensiveness.
Ans. Start at the top right and do a staircase search in O(m+n) time and O(1) space. Compare the current value with the target. If it is equal, return found. If it is larger, move left. If it is smaller, move down. Trace repeats until found or indices leave the matrix.
Ans. I choose the chart based on the data type and the question being asked. Use a bar chart for comparing categories, a histogram for showing the distribution of one numeric variable, a box plot for spread and outliers across groups, and a line chart for trends over time or ordered sequences.
Ans. SQL retrieves and manages relational data: SELECT reads rows, JOIN combines related tables, primary keys uniquely identify rows, foreign keys enforce relationships, normalisation reduces duplication, and ACID makes transactions reliable. The key detail is integrity: constraints and normal forms keep data consistent, while atomicity, consistency, isolation, and durability protect changes during failures and concurrency.
Ans. Use a queue for normal breadth first traversal and a stack to reverse the output order. Enqueue the root, then repeatedly dequeue a node, push it onto the stack, enqueue its right child first and left child second. Finally pop all stack items to print bottom-up left-to-right. Time is O(n), space is O(n).
Ans. Work backwards from the destination to the source. Since each forward move adds one coordinate to the other, the reverse move subtracts the smaller coordinate from the larger. Repeated subtraction can be optimised with modulo. If you reach exactly (x1, y1), print “Yes”; if any coordinate becomes smaller than the source, print “No”. Time is logarithmic.
Ans. Iterate through the tuple and dispatch by type using a dictionary that maps list to a function printing its items with newline separators, and str to a function printing the string directly. Use type(element) to select the action, avoiding if statements. The time complexity is O(n + m), where m is printed list items.
Ans. Model Instagram around User, Post, Media, Follow, Like, Comment and Feed objects, backed by separate services for identity, posting, social graph, engagement, media storage and feed generation. The key detail is feed fan-out: push new posts to follower feed caches for normal users, but pull and rank posts on read for celebrities.
Ans. Reverse a linked list by iterating through it and changing each node’s next pointer to point to the previous node. Keep three pointers: previous, current, and next, so you do not lose the rest of the list. At the end, previous is the new head. Time complexity is O(n), space complexity is O(1).
Ans. Identify the question type first, then write down the given data clearly. Convert words into equations, ratios, tables, or diagrams as needed. Use shortcuts only when they are reliable. For reasoning, look for patterns, conditions, and eliminations. Check the final answer against the question to avoid calculation or interpretation errors.
No question matches those filters. Reset filters.
Showing 60 of 149 questions. Ranked by how often the same question came back across interviews.
A spoken interview built from these questions, scored when you finish; the feedback is yours.
Start practisingWhen you are ready, record The One: a single interview hiring teams watch, so you stop repeating first rounds.
Candidate interviews most often cover CS fundamentals (60%) and DSA (30%).
Candidate interviews show an average of 2.9 rounds per experience, with a typical sequence of Online test → Technical → HR. Individual interview paths can vary.
Among questions with a recorded difficulty, the mix is easy 58%, medium 40%, hard 3%.
