Oracle interview notes, page 3 of 9

Large Enterprises

Most of these Oracle write-ups come from Software Engineers (83), then Machine Learning Engineers (2) and Data Scientists (1). The most reported rounds are the technical phone screen (52), the system design round (14), the behavioral / leadership round (12) and the hiring manager screen (3).

87 interview notes · updated Oct 2026

Oracle·Software Engineer·Onsite - Behavioral / Leadership

May 2026

Behavioral round at Oracle for a software engineer role, pretty much one meaty question that took up most of the conversation.

Tell me about a time you took on work that was clearly outside your role to push a project forward. What made it out of scope, what did you risk, how did you get the actual owners on board, and what happened after you handed it back?

“This one has a lot of moving parts and I underestimated that going in.”

View Post

Oracle·Software Engineer·Technical Phone Screen

May 2026

Oracle SWE interview that came down to a classic cache design problem. Nothing too surprising but the implementation details are where it gets you.

Implement an LRU cache that supports get and put operations in O(1) time, evicting the least recently used entry when the cache is full.

“I knew the answer involved a hash map and a doubly linked list but explaining why you need both at the same time, while also coding it up, is harder than it sounds.”

View Post

Oracle·Software Engineer·Onsite - Behavioral / Leadership

May 2026

Behavioral screen at Oracle for a software engineer role. Two questions, both pretty standard but the second one is sneakier than it looks if you haven't thought about it before.

Walk me through one or two past projects, then pick the most technically challenging problem you solved. What was the goal, what did you personally do, what trade-offs did you make, and what was the measurable outcome?
How do you approach a problem you've never seen before and don't have enough information to solve yet? Walk me through how you'd get oriented, fill in the gaps, make progress without getting blocked, and keep stakeholders in the loop.

“I had a project ready but fumbled the trade-offs part.”

View Post

Oracle·Software Engineer·Technical Phone Screen

May 2026

Oracle SWE interview where the main event was a deep project walkthrough. They weren't interested in the high-level story, they kept drilling down until you either knew your stuff cold or ran out of answers.

Walk me through a project you worked on in detail, including your specific contributions, the technical decisions you made, trade-offs involved, challenges you ran into, and what the actual impact was.

“This sounds like a normal project question until they start pulling threads.”

View Post

Oracle·Software Engineer·Onsite - Behavioral / Leadership

May 2026

Two behavioral questions for a software engineer role at Oracle. Pretty standard stuff but the second one tripped me up more than I expected.

Tell me about a time you changed the scope of a project. What drove the change, how did you get stakeholders on board, and what happened in the end?
Describe a process you automated. What was the painful manual part, what did you build to fix it, and how did you measure the improvement?

“I had a decent story for this but fumbled the stakeholder alignment part.”

View Post

Oracle·Software Engineer·Technical Phone Screen

May 2026

Oracle SWE interview with a string manipulation problem that looked easy but had a few gotchas worth thinking through beforehand.

Given a list of phrases and a list of words, for each phrase find the 0-based index of the first word in that phrase that is an anagram of any word in the list. Return -1 for phrases with no match.

“My first instinct was to sort the characters of every candidate word and stick them in a set, then do the same for each word in each phrase as I scanned through.”

View Post

Oracle·Software Engineer·Technical Phone Screen

May 2026

Oracle SWE coding round, one question the whole session. Pretty much a log parsing problem that looks easy until you actually think through the edge cases.

Given a list of unsorted log entries in the format 'userId action timestamp' (where action is either signin or signout), and an integer maxTime, return all userIds that have at least one valid session. A session is a matched signin/signout pair for the same user where the duration is at most maxTime. Unmatched signins or signouts should be ignored.

“Spent the first few minutes just restating the problem back to make sure I understood the pairing logic.”

View Post

Oracle·Software Engineer·Technical Phone Screen

May 2026

Oracle SWE interview with a string decoding problem that looks manageable until you actually sit down and think through all the edge cases. Single coding question, parsing-heavy, and the O(1) space constraint is what makes it interesting.

You're given a specially encoded string where letters a-i map to digits 1-9, letters j-z map to two digits followed by a '#', and consecutive repeated characters are compressed with a count in parentheses. Write a function that decodes the string and returns an array of 26 integers representing how many times each letter appears.

“The encoding rules themselves aren't hard to follow but parsing everything in one pass without building the decoded string is where I got tripped up.”

View Post

Oracle·Software Engineer·Technical Phone Screen

May 2026

Oracle SWE coding round, one algorithmic question the whole time. The interviewer was pretty focused on complexity analysis afterward which I wasn't fully ready for.

Given an integer n, find the largest number less than or equal to n whose digits are monotone increasing (each digit is less than or equal to the next).

“I got the greedy idea pretty quickly.”

View Post

Oracle·Software Engineer·Onsite - System Design / Architecture

May 2026

Oracle system design round for a software engineering role. The whole thing was basically one massive question about deploying a generative AI model at scale, and they wanted you to cover pretty much every layer of the stack. Dense.

Design a production deployment for a generative AI model that needs to handle high traffic and stay available. Walk through the inference serving setup, how requests get routed, autoscaling with GPU scheduling, multi-region failover, model versioning and rollout strategy, managing stateful dependencies like tokenizers and embedding caches, observability, rate limiting, and how you'd hit latency and throughput SLOs during traffic spikes or partial failures.

“This was a lot.”

View Post