Adobe runs an unusually long, elimination-style engineering loop that is heavily weighted toward data structures and algorithms plus core CS fundamentals, with comparatively little classic distributed-systems design until senior levels. Campus and intern hiring in India (Noida/Bengaluru) is compressed into an online assessment of aptitude and CS-fundamental MCQs plus two coding problems, followed by usually one or two technical interviews. Experienced hiring for MTS/Computer Scientist levels stacks three to seven rounds - multiple DSA rounds, a language/OS deep-dive, low-level design, a hiring-manager round and often a director round - with candidates advancing only after clearing each one.
These 19 writeups cover software engineering, frontend engineering, internship and other roles, India-heavy (Noida), some remote/US. 16 of them were posted under a pseudonym (Reddit, GeeksforGeeks or forum handles), so we could not verify the authors’ identities.
Typical rounds
3.51-7 range
Outcomes shared
8/3offer / not
Most common round
Technical
Sources span
2022 - 2025
Most frequently reported · Data structures & algorithms(17), Arrays & strings(12), Past projects(12), OOP(11), Trees(11)
Adobe’s official process
Adobe's own careers site describes a five-step hiring flow: create a candidate profile and apply (it explicitly invites a portfolio or GitHub link for creative and technical roles), a conversation with a Talent Partner about background and the role, an interview with the hiring manager focused on problem-solving and fit, skills assessments plus team interviews, and then offer/onboarding. The page also warns that genuine Adobe recruiting only comes from adobe.com addresses and gives per-step preparation tips, but it does not publish round counts or formats for specific job families.
A pure paper screen on the resume before anyone was invited to the assessment.
Covered · Resume deep-dive
2
Online assessmentOnline assessment
Thirty multiple-choice questions spread across aptitude and CS fundamentals, most of them testing basics rather than depth, plus two coding problems on linked lists and arrays.
The author notes the round varied a lot between candidates: some got only resume questions, others only coding. This candidate got four easy-to-moderate problems spanning array frequency counting, matrix traversal, tree level-order and graph BFS.
Covered · Arrays & strings, Trees, Graphs, Past projects
A mixed aptitude section covering logical, quantitative and verbal reasoning, followed by two algorithmic coding questions.
Covered · Aptitude, Data structures & algorithms
2
Technical interviewTechnical
Started as a resume-driven conversation, including a deep dive into a cybersecurity hackathon the candidate had listed, then moved through core-subject theory and two coding problems on circular linked lists and a house-robber style dynamic programming task.
Covered · Resume deep-dive, OOP, DBMS, Networking, Past projects
Twenty-five MCQs on core subjects with Java-flavoured OOP questions, plus two straightforward coding problems. Shortlisting before this was purely CGPA-based.
Covered · Operating systems, DBMS, OOP, Data structures & algorithms
2
Technical interviewTechnical
Three problems: a rat-in-a-maze path enumeration, writing quicksort correctly end to end, and building and querying a prefix-sum array.
Two coding problems ranging from easy-medium to medium-hard alongside thirty MCQs on C, DBMS and OOP. The author only got full test-case acceptance in the closing minutes.
Covered · Data structures & algorithms, DBMS, OOP
2
Technical interviewTechnical
Opened with a project walkthrough including a live demo that failed on the spot, then four coding problems on a shared CodePair editor with built-in test cases: merging sorted arrays, minimum swaps to balance parentheses, implementing a queue from stacks, and finding a linked list's midpoint. Some were code, some approach-only.
Covered · Past projects, Arrays & strings, Stacks and queues, Linked lists
Thirty-seven questions in two hours, of which thirty-five were MCQs and two were coding problems.
Covered · Data structures & algorithms, Aptitude
2
Technical interviewTechnical
A single-project discussion followed by a hard binary-tree traversal problem and easier binary-search-tree questions. The candidate had the approach but was faulted for not covering edge cases and complexity.
Medium-difficulty DSA coding problems plus theory MCQs on data structures and OOP. Thirty-five candidates were shortlisted from the whole cohort.
Covered · Data structures & algorithms, OOP
2
Technical interviewTechnical
Roughly fifty minutes, with about twenty of them spent dissecting a web-development project. Then an easy array-scanning problem solved with a dry run and complexity discussion, followed by a binary-tree path problem where the interviewer pushed for a lower-space optimisation. Closed by asking why the candidate wanted Adobe.
An emailed online assessment after applying through a diversity-focused drive whose link was routed via the college placement cell; clearing it triggered scheduled interview rounds.
Covered · Data structures & algorithms, OOP, Puzzles
2
Interview roundsTechnical
The author describes scheduled interviews testing both technical and problem-solving ability roughly a week before the final decision, but does not break down individual rounds.
Covered · Data structures & algorithms, Past projects
Ninety minutes total: an hour for two basic coding problems and then three ten-minute aptitude sections covering mental ability, comprehension and reasoning.
Covered · Data structures & algorithms, Aptitude
2
Technical interviewTechnical
A single binary-tree boundary-traversal problem, with the interviewer specifically pushing for a one-pass solution after the approach was explained.
Covered · Trees
3
Technical interviewTechnical
Two problems: a sliding-window string question with a bounded number of allowed character changes, and a grid shortest-path question with a budget for removing obstacles, which was then made harder by allowing movement in all four directions.
A blend of project discussion, HR-style questions and computer-networking theory, closing with a linked-list rearrangement problem separating odd and even positioned nodes.
Covered · Past projects, Behavioral, Networking, Linked lists
An hour of DSA where the interviewer expected production-quality code, covering fast exponentiation including negative exponents, search in a sorted matrix, and grouping anagrams.
Scheduled with a different team's manager because the previous round's feedback was only average. Nearly the whole hour went on one optimisation problem, taken from a recursive solution to a DP formulation.
Covered · Dynamic programming
5
Director roundFinal
Mostly behavioural: current role, what the candidate personally built, challenges, how a manager would describe them and where they rank in their team, each answer needing justification. A Dutch-national-flag sorting problem was slotted in the middle, coded in an IDE of the candidate's choice.
Covered · Behavioral, Past projects, Arrays & strings
A problem-solving hour: integer square root without library functions, array rotation with a demand for a divide-and-conquer alternative, minimum squares summing to n, connecting same-level tree nodes in constant space, and a box-stacking maximisation with fixed then variable dimensions.
Because the candidate was not strong in Java, this round went deep on Java concepts: singleton and thread-safe initialisation, garbage collection, the string pool and immutability, inheritance versus composition, dependency injection, HashMap internals, equals/hashCode contracts, strategy and observer patterns, and thread coordination.
Roughly fifty minutes: spiral tree traversal with a push to use one stack instead of two, recursively removing adjacent duplicate characters with an optimisation hint, string rotation checking, polymorphism types and building an immutable class.
Covered · Trees, Arrays & strings, OOP
4
Director roundFinal
Spiral printing of a rectangular matrix, maximum interval overlap, enumerating ways a Java program can crash and the exception each throws, and reasoning about output of a reference-reassignment snippet.
Two hours with fifteen aptitude MCQs and two coding questions, run through a third-party partner. Candidates already placed at high-paying firms skipped straight to interviews.
Covered · Aptitude, Data structures & algorithms
2
Technical interviewTechnical
The longest round at eighty to ninety minutes and explicitly used to eliminate most candidates. Covered a house-robber DP with pseudo code, a next-greater-element variant fully coded, external sorting of a file larger than RAM, timing a spreading process across a tree, and a large phone-directory search design where follow-ups pushed towards indexed lookup structures. Ended with OS and OOP theory.
Seventy-five minutes for twenty-two questions: fifteen logical-reasoning items, five C++ questions and two coding problems on tallying votes and on ordered substring selection.
Covered · Aptitude, Data structures & algorithms, OOP
2
Technical interviewTechnical
K-largest elements, running median over a stream, and a word-ladder style shortest transformation using a dictionary.
Covered · Heaps, Graphs, Data structures & algorithms
3
Technical interviewTechnical
The circular house-robber variant and its linear form, plus computing the minimum number of meeting rooms from meeting intervals.
Systems-flavoured C: dynamic one- and two-dimensional array allocation with single and double pointers, implementing memcpy and strcmp, a discussion of threads, locks, deadlocks and endianness, and spiral matrix printing.
Two open-ended modelling problems: whether a paragraph can be spelled from a set of lettered cubes, with a mirror-letter twist, and whether a road can be crossed given circular potholes described by centre and radius.
Covered · Data structures & algorithms, Puzzles, Geometry
6
Managerial roundHiring manager
Designing an elevator plus its dispatch algorithm, fitting a binary search tree into a fixed hundred-byte arena, handling the resulting internal fragmentation after deletions, and optimising a case study drawn from the manager's own team.
Covered · Low-level design, Memory management, Case study
7
Director roundFinal
Reasoning about two processes touching the same raw address and why the second fails, a walk through paging, address translation and thrashing, and search/insert/delete on a sorted two-dimensional array. The offer call came about ten days later.
Covered · Operating systems, Debugging, Data structures & algorithms
Printing ancestors in a binary tree, finding the intersection of two linked lists, a factory-pattern question, C++ smart pointers, and a code-hygiene question about resetting global state across many return paths.
Dynamic allocation and deallocation of a two-dimensional array, iterative pre-order traversal, reversing a linked list, and C++ virtual and override semantics.
Designing a thread-safe chat box in C++ with mutexes and condition variables, plus product-adjacent questions on LRU caching and image scaling and translation, reflecting the Photoshop team the candidate was interviewing for.
Two forty-five minute sections on HackerRank: an objective aptitude test and a language test in C/C++ or Java mixing objective output questions with short coding tasks such as base conversion and generating a unique number set.
Covered · Aptitude, Data structures & algorithms, Debugging
2
Technical interviewTechnical
Palindromic substring enumeration, deadlock and virtual-memory theory followed by coding a virtual-memory scheme, and a diagnosis scenario about a browser-cache-database path whose latency jumped after a bug fix.
Base conversion, then generalising and coding a concentric-square matrix pattern, then an open conversation about Adobe's own products and Creative Cloud.
Covered · Arrays & strings, Domain knowledge
4
Technical interviewTechnical
Two classic optimisation puzzles, zig-zag level order traversal, loop detection and removal in a linked list, and Windows internals questions on the hosts file, the loopback address, the registry and 32- versus 64-bit differences.
Covered · Puzzles, Trees, Linked lists, Operating systems
5
Senior interviewFinal
With the most senior interviewer: a project deep dive on practical usage, then open-ended scenario questions such as debugging a dead wireless keyboard, how to verify a blur effect really applied to an image, and proposing missing features for consumer apps.
Covered · Past projects, Product sense, Customer scenarios
Resume and performance-optimisation discussion, a deep dive on linked-list use cases and LRU cache, and which sorting algorithm suits which scenario, then two array problems with heavy emphasis on complexity.
Two more standard algorithm problems: grouping anagrams and a stack-simulation collision problem.
Covered · Arrays & strings, Stacks and queues
3
Advanced JavaScriptTechnical
Critical rendering path, React memoisation hooks and the diffing algorithm, a prototypal-inheritance snippet, and building a task runner that executes promises in batches mixing parallel and sequential resolution.
Designing a file-directory system, with the discussion pushed into component structure, API contracts, scalability, performance and list virtualisation.
Two problems in a single-day hiring drive: counting sentence variants formed by anagram substitution, and finding the centroid nodes of a tree modelled as a company reporting hierarchy. The next-round invitation arrived within ten minutes.
Covered · Arrays & strings, Trees, Graphs
2
Problem solvingTechnical
A repeat of the tree-centroid problem with a complexity target, plus a top-k frequent elements problem framed as an e-commerce recommendation task.
Covered · Trees, Heaps, Data structures & algorithms
3
Low level designLow-level design
Designing a notification service that fans out to SMS, WhatsApp and push channels, covering design-pattern choice, class-level structure, a diagram, and how user interaction with notifications would be tracked.
Covered · Low-level design, Design patterns, OOP
4
Hiring managerHiring manager
Microservices-versus-monolith justification for the candidate's own architecture, a disagreement-with-management STAR question, and an unexpected object-modelling exercise: model a water bottle using encapsulation and inheritance.
JavaScript fundamentals probed through scenarios rather than definitions - event loop behaviour, closures - plus React basics and two algorithm problems on trees and array manipulation.
How cache invalidation works in the candidate's product and how it is served to millions of users end to end from deployment to device, then frontend performance optimisations used in practice, then a medium tree problem.
Covered · Scalability, Caching, Trees
3
Hiring managerHiring manager
A project the candidate was proudest of and why, then two coded problems: bottom view of a binary tree from scratch and a binary-search-on-answer problem.
Covered · Past projects, Trees, Binary search
4
Senior hiring managerHiring manager
A React machine-coding exercise, a spatial reasoning puzzle about cutting and painting a cube, and one more medium tree problem.
Covered · React, Puzzles, Trees
5
Design and architectureSystem design
Not a generic design-a-product round: the candidate had to draw the high-level design of their own project and defend each decision, discuss avoiding a utility library and reimplement one of its functions, and print a Fibonacci sequence without storing values.
Covered · System design, Past projects, API design
About seventy minutes: running median with two heaps and an LRU cache, both pseudo code only, wrapped around broad theory on scaling, load balancers, CAP, locking, SQL versus NoSQL, normalisation and atomicity.
Covered · Heaps, System design, DBMS, Operating systems, Concurrency
2
Technical interviewTechnical
Two fully coded and test-run problems, one a long constraint-heavy backtracking question, followed by experience-driven questions on CI/CD, build tooling, cloud, containers and streaming infrastructure.
Covered · Backtracking, Data structures & algorithms, DevOps, Scalability
3
Hiring managerHiring manager
A grilling on ownership, end-to-end delivery, leadership and debugging, and the point at which the candidate learned the role was release-engineering focused. Then scheduling problems: minimising makespan across two workers, and handling job dependencies via topological ordering.
Covered · Behavioral, Past projects, Greedy, Graphs
4
Director roundFinal
The candidate cleared everything and was invited to a director round described as another hiring-manager-style conversation, but declined it after accepting a competing offer.
An hour opening with introductions and an overview of the candidate's current work and tech stack before moving into algorithm discussion. The remainder of the writeup sits behind Medium's member wall.
Covered · Data structures & algorithms, Resume deep-dive
Adobe runs an unusually long, elimination-style engineering loop that is heavily weighted toward data structures and algorithms plus core CS fundamentals, with comparatively little classic distributed-systems design until senior levels. Campus and intern hiring in India (Noida/Bengaluru) is compressed into an online assessment of aptitude and CS-fundamental MCQs plus two coding problems, followed by usually one or two technical interviews. Experienced hiring for MTS/Computer Scientist levels stacks three to seven rounds - multiple DSA rounds, a language/OS deep-dive, low-level design, a hiring-manager round and often a director round - with candidates advancing only after clearing each one.
What topics does Adobe test in interviews?
Commonly reported topics include Data structures & algorithms, Trees, Arrays & strings, Dynamic programming, Graphs, Low-level design.
These guides summarize public, first-hand interview experiences shared by candidates. They describe the shape of each loop and the topics that came up, not a leaked question bank, and every experience links back to its original source. Processes change often; treat this as directional prep, not a script.