← Back to Directory

Adobe interview notes, page 5 of 6

Large Enterprises

Most of these Adobe write-ups come from Software Engineers (30), then Machine Learning Engineers (15) and Product Managers (8). The most reported rounds are the technical phone screen (32), the system design round (8), the product sense round (7) and the multi-round onsite (4).

58 interview notes · updated Sep 2026

Adobe·Software Engineer·Onsite - Behavioral / Leadership

Jun 2026
Went through leadership-focused rounds at Adobe for a software engineering role and got rejected. No technical questions at all, just vague trust-building conversations that I couldn't figure out how to navigate.
  • How would you earn the trust of your manager and your team?

“This is the question that got me both times.”

View Post

Adobe·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Product strategy question for what seemed like a senior PM role, asked to build a three-year innovation roadmap for Chrome on large screens. Pretty open-ended, which I wasn't fully prepared for.
  • You're a PM for Google Chrome on large screens. Build a three-year innovation roadmap.

“I spent too long on the current-state analysis and didn't leave enough room to actually talk through the roadmap years in any meaningful way.”

View Post

Adobe·Software Engineer·Technical Phone Screen

Jun 2026
Adobe coding round focused entirely on task scheduling with dependency graphs. Pretty classic topological sort territory but they pushed hard on complexity analysis and edge case behavior, which is where things got interesting.
  • Given n tasks labeled 0 to n-1 and a list of dependency pairs where (a, b) means task a depends on task b, write a function to determine if all tasks can be completed, and if so, return a valid execution order.

“I knew this was topological sort pretty much immediately, but I fumbled the cycle detection explanation for a bit.”

View Post

Adobe·Software Engineer·Technical Phone Screen

Jun 2026
Adobe technical screen for a software engineer role, basically one big Java Streams question broken into four parts. The whole thing felt like a functional programming deep-dive and I was not fully prepared for the Collector API stuff.
  • Given a Student class with name, grade, and subject fields, implement a method that returns a Collector grouping students by subject and computing the average grade per subject using Java Streams.
  • Implement a method that returns a Predicate<Student> filtering students in a given subject whose grade exceeds a specified threshold.
  • Implement a method that returns a Stream<Student> of the top N students in a given subject, sorted by grade in descending order.
  • Implement a method that returns a Collector computing the median grade per subject, using idiomatic Stream and Collector pipelines without any imperative loops.

“This one I got through okay.”

View Post

Adobe·Data Scientist·Technical Phone Screen

Jun 2026
Adobe DS interview with a product analytics case built around Adobe Express. One meaty scenario question that spiraled into a bunch of sub-topics. Felt like a structured case more than a pure technical screen.
  • Adobe Express is seeing around 1 million downloads per day, but daily active users aren't growing. How would you investigate this gap, and what analyses, hypotheses, and experiments would you run?

“This one sprawled more than I expected.”

View Post

Adobe·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Adobe ML engineer interview that went pretty deep into inference optimization, specifically the KV cache. Not the most brutal round I've had but you need to actually understand the mechanics, not just know the term.
  • How does the KV cache work in transformer inference, and why do we cache keys and values but not queries?
  • Why does KV caching matter specifically for decoder-only models, and why does it not provide the same benefit for encoder architectures?
  • What are the memory tradeoffs involved in KV caching as sequence length grows?

“This is where I started strong and then slightly fumbled the Q part.”

View Post

Adobe·Software Engineer·Technical Phone Screen

Jun 2026
Adobe software engineer interview where the whole thing basically revolves around your resume. They pick something you listed and just start pulling threads until you run out of answers.
  • Walk me through your resume, and be ready to go deep on anything you mention.

“This is deceptively stressful.”

View Post

Adobe·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Adobe SWE interview with a single system design question about building type-ahead search for a large e-commerce site. Pretty focused session, no fluff.
  • How would you design a type-ahead (autocomplete) search feature for a large e-commerce platform like walmart.com?

“Spent probably too long on the trie data structure angle before they nudged me toward scale considerations.”

View Post

Adobe·Software Engineer·Technical Phone Screen

Jun 2026
Adobe SWE interview that went deep on a file processing problem. The question started simple but kept expanding into memory constraints and data structure trade-offs, which I wasn't fully ready for.
  • Write a function to read a large text file and return the frequency count of each word. You'll need to define your tokenization and normalization strategy, handle memory constraints through streaming or chunking, and return the top-k most frequent words. Also analyze time and space complexity and compare hash maps, tries, and external sorting when the file barely fits in memory.

“This one kept growing legs.”

View Post

Adobe·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Adobe ML engineer interview that was pretty much a loss function quiz. Three questions, all theory-heavy, felt like a written exam more than a conversation.
  • What loss function is used in logistic regression, and what's the reasoning behind that choice?
  • What loss function does linear regression use, and why is it the standard choice?
  • Compare MSE and MAE across their mathematical forms, how they handle outliers, gradient behavior, and when you'd pick one over the other.

“Binary cross-entropy, and the 'why' is the part people fumble.”

View Post