Microsoft Interview Questions — Page 59

Microsoft·Software Engineer·Onsite - Multi Round

Jul 2026

Three rounds at Microsoft Dublin for a software engineering role. The first two went fine with supportive interviewers, but the final round with a principal manager was rough and I left with no idea where I stood.

Write a linked list implementation from scratch, including the node class, then implement a function to delete the node with the maximum value. You also need to construct your own test input and print the output.

“This is where things fell apart.”

View Post

Microsoft·Software Engineer·Technical Phone Screen

Jul 2026

Microsoft SWE interview with a classic k-way merge problem. Pretty standard algorithmic round but the constraints pushed you toward actually knowing your heap usage rather than brute-forcing it.

Given k sorted arrays with a total of N elements, merge them into a single sorted array using an asymptotically optimal algorithm. Describe your approach including data structures, time and space complexity, then implement it.

“The naive approach of scanning all k heads at each step is O(N*k) and they basically told you upfront that wasn't acceptable.”

View Post

Microsoft·Product Manager·Onsite - Behavioral / Leadership

Jul 2026

Went through a behavioral round for a PM role at Microsoft. Pretty standard set of questions but a couple of them pushed me to think harder than I expected. Nothing too wild, but worth writing up.

Tell me about a time you had to make a quick decision without having all the information you needed.
What project are you most proud of, and what makes it stand out for you?
Describe the hardest challenge you've faced professionally and walk me through how you resolved it.
Can you give an example of a time you disagreed with your manager or a peer? How did you handle it?
Walk me through a time you improved or optimized a process within your team or project.

“I had a decent story ready for this but fumbled the landing.”

View Post

Microsoft·Data Scientist·Onsite - Behavioral / Leadership

Jul 2026

Microsoft Data Scientist onsite focused heavily on a single deep behavioral question about owning an ML project from scratch. The whole thing was structured around ambiguity and leadership, less about coding and more about whether you can actually run something end to end.

Walk me through a time you led an ML project end to end where the requirements were unclear. How did you define the problem, choose your approach, align stakeholders, and measure success?

“This question is doing a lot of work at once and I underestimated that going in.”

View Post

Microsoft·Software Engineer·Technical Phone Screen

Jul 2026

Microsoft SWE interview with a grid-based BFS problem. Pretty standard algorithmic round but the problem had enough edge cases to keep you honest.

You're given a 2D grid where cells are either walls, gates, or empty rooms. Fill each empty room with its shortest distance to any gate, moving only up/down/left/right. Walls and unreachable rooms stay as-is.

“My first instinct was to BFS from every empty room individually and find the nearest gate.”

View Post

Microsoft·Software Engineer·Technical Phone Screen

Jul 2026

Microsoft SWE interview with a graph/tree problem that looks deceptively clean until you actually try to implement it. One question, but it had enough edge cases to keep me busy.

You're given a list of directed edges that originally formed a rooted tree, but one extra edge was added. Find and return the edge to remove so the graph becomes a valid rooted tree again. If multiple candidates exist, return the last one in the input.

“I knew union-find was involved but kept second-guessing myself on the two-case split: when a node ends up with two parents vs when there's a cycle but no double-parent.”

View Post

Microsoft·Software Engineer·Technical Phone Screen

Jul 2026

Microsoft SWE interview with a string assembly problem that escalated across three parts. Each follow-up added a new wrinkle and by Q3 I was scrambling to keep my graph logic clean.

You have a list of fragments, each with a start tag, end tag, and payload string. Fragments chain together when one fragment's end tag matches the next fragment's start tag. Given that the fragments form exactly one continuous chain, assemble and return the concatenated payload string.
Now treat each fragment as bidirectional. You can traverse the chain forward or backward, and either ordering of the payloads is a valid answer. Modify your solution to handle this.
The input may now contain multiple disjoint chains. Each group of fragments forms its own independent chain with no branching. Return all assembled strings, one per chain.

“Felt straightforward at first.”

View Post

Microsoft·Software Engineer·Technical Phone Screen

Jul 2026

Phone screen for a Software Engineer role at Microsoft where they handed me an AI-written URL shortener and asked me to tear it apart. Decent question, though I'd say the follow-ups were where it got genuinely hard.

You're given an AI-generated in-memory URL shortener implementation. Walk through the bugs and correctness issues in both the shorten and resolve methods, explain why each is a problem, and describe how you'd fix them.
How would you write a unit test that deterministically reproduces the collision bug in the code generator?
If this service had to run across multiple processes or servers, which assumptions in the current in-memory design break and what would you change?
How would you make the code generation both collision-resistant and unguessable at the same time?
If the url_to_code reverse map was removed entirely, what behavior changes and does that fix or remove any of the existing bugs?

“The collision bug in shorten is the obvious one, and I spotted it pretty quick.”

View Post

Microsoft·Data Scientist·Onsite - Multi Round

Jul 2026

Microsoft Data Scientist interview that went deep on the full ML lifecycle, from problem framing all the way through post-launch monitoring. It was a lot to cover in one session and I felt like I was speed-running my entire career.

Walk me through an ML project you owned end-to-end: how did you frame the problem, define success metrics tied to business KPIs, and get alignment across PM, Engineering, and Legal?
How did you handle data sourcing and privacy constraints in that project?
How did you choose your model and evaluate it both offline and online? What did your experiment design look like?
Tell me about a time a PM strongly pushed back on your recommendation. How did you change their mind, and what was the measurable outcome?
Describe a failure in an ML project: what broke, what did you change, and how did you make the process more reproducible afterward?
How do you approach fairness and ethical considerations in a model when you're under real time pressure to ship?

“This is the kind of question that sounds manageable until you realize they want specifics on every single handoff.”

View Post

Microsoft·Product Manager·Onsite - System Design / Architecture

Jul 2026

Microsoft PM interview with a pretty technical system design angle. The question was about diagnosing a broken ML pipeline, which felt more like a data engineering interview than a product role, but I get why they ask it.

Your time-series ML pipeline has become unstable and prediction accuracy is degrading. Walk through how you'd diagnose the root cause across the data ingestion, feature generation, and model-serving layers. What evaluation framework, sampling strategy, and metrics would you use to quantify the regression? And what technical and process changes would you make to restore and maintain stability?

“This one is broad in a sneaky way.”

View Post