Anthropic Interview Questions — Page 30

Anthropic·Software Engineer·Technical Phone Screen

Jul 2026

Anthropic SWE interview that was basically one long design problem about building a text classifier on top of a black-box LLM scoring API. Dense question, a lot of moving parts, and they clearly wanted to see how far you could push the design before you ran out of ideas.

Given a helper function that returns per-token probabilities from an LLM, design a complete binary text classifier: construct prompts for each class, produce a continuous probability score per input, handle numerical stability with log-probabilities, aggregate token-level scores to sequence level, define batching, select thresholds, evaluate with ROC-AUC and F1, propose improvements like prompt ensembling and calibration, discuss failure modes, and give pseudocode with complexity analysis.

“This was one question but it kept expanding.”

View Post

Anthropic·Software Engineer·Technical Phone Screen

Jul 2026

Anthropic software engineering interview with a coding question centered on implementing an LRU-based memoization helper from scratch, plus a persistence/recovery follow-up that caught me a bit flat-footed.

Implement an LRU memoization helper class with a configurable capacity. It should include a method to generate a deterministic cache key from a function and its arguments (handling both positional and keyword args, with keyword arg order not mattering), and a method that returns a cached result if available or computes, caches, and returns it otherwise.
Follow-up: if the process crashes and the in-memory cache is lost, how would you persist the cache to disk so it can be restored correctly after a restart? What data would you write, when would you write it, and how would recovery work?

“I knew the LRU eviction part pretty well, ordered dict does most of the heavy lifting there.”

View Post

Anthropic·Software Engineer·Onsite - System Design / Architecture

Jul 2026

Anthropic system design round, got a routing problem that looked deceptively simple at first glance but kept expanding. The follow-up questions about consistency and load were where things got interesting.

You have multiple GPT servers and a stream of incoming prompts. Design a routing system using a hash-table-based approach so that the same prompt always goes to the same server. Implement add_server, remove_server, and route(prompt), and walk through your data structure choices. Then extend it to handle load balancing and what happens when servers are added or removed.

“Started with the obvious stuff, md5 the prompt, mod by server count, done.”

View Post

Anthropic·Software Engineer·Technical Phone Screen

Jul 2026

Anthropic SWE interview that was more hands-on than I expected. The coding part was an image processing pipeline, which felt a bit niche, and then the follow-up spiraled into a full distributed systems conversation. Not a bad experience but definitely came in underprepared for the depth of the system design portion.

Implement an image processing pipeline using a library like PIL or OpenCV. Utility functions for file I/O are provided, so focus on the actual image operations: resize, rotate, grayscale, blur, format conversion, etc.
How would you parallelize this pipeline to handle many images at once, and how would you scale it beyond a single machine?

“I spent way too long second-guessing whether to use PIL or OpenCV and ended up picking PIL which was fine.”

View Post

Anthropic·Software Engineer·Onsite - System Design / Architecture

Jul 2026

Interviewed for a software engineering role at Anthropic and got a system design question that was more nuanced than I expected. The versioning angle made it trickier than a plain CRUD task list problem.

Design a system that stores a user's task list at specific points in time and allows querying what that list looked like at any arbitrary past timestamp.

“The point-in-time query part is what trips you up if you go in thinking it's just a task list.”

View Post

Anthropic·Software Engineer·Onsite - Behavioral / Leadership

Jul 2026

Interviewed for a PMM role at Anthropic, got a behavioral question about adapting a campaign on the fly. Pretty standard round but the question had more layers than I expected.

Describe a situation where you had to change your marketing strategy in the middle of an active campaign.

“I had a decent story ready but I kept second-guessing whether my pivot was dramatic enough to impress them.”

View Post

Anthropic·Machine Learning Engineer·Technical Phone Screen

Jul 2026

Coding screen for an MLE role at Anthropic. The problem was deceptively implementation-heavy for what sounds like a simple trace-parsing task, and the follow-up added a filtering layer that I didn't fully think through in time.

Given a sequence of function-name samples from a profiler (one per tick), convert it into a list of events where each event captures a function name and its [start, end) time interval. Consecutive samples of the same function belong to one event; a new function starts a new event.
Follow-up: modify your solution so that only events where a function appears for at least k consecutive samples are emitted. Shorter runs should be silently discarded.

“Sounds trivial until you're actually writing it out.”

View Post

Anthropic·Software Engineer·Onsite - System Design / Architecture

Jul 2026

System design round at Anthropic for a software engineering role. The whole question was about deduplicating files across a massive directory tree, which sounds manageable until you start thinking about scale and the interviewer keeps asking follow-ups.

Design a program to find and deduplicate files across a very large directory tree, without loading entire files into memory. Walk through your hashing strategy, how you handle very large files, hash collisions, hard-linking duplicates safely, and how you'd parallelize the whole thing across cores or machines.

“This question has a lot of layers and I kept getting pulled into the wrong ones first.”

View Post

Anthropic·Machine Learning Engineer·Technical Phone Screen

Jul 2026

Interviewed for an ML engineer role at Anthropic. One technical question about controlling model creativity, which sounds deceptively simple but actually has a lot of depth to it once you get into the weeds.

How do you control or adjust the creativity of generative AI models?

“Went straight to temperature and top-p sampling, which is the obvious answer, but then they pushed further and I realized I hadn't thought carefully about the tradeoffs between them.”

View Post

Anthropic·Software Engineer·Technical Phone Screen

Jul 2026

Anthropic software engineer interview with a classic systems/algorithms question dressed up as a real product scenario. The crawler problem sounds approachable until you're actually in it and realize they want you to think about everything at once.

Build a single-threaded web crawler that starts from a given URL and collects all unique pages reachable within the same hostname. You're given a helper function that returns all hyperlinks on a page. Walk through your design choices: BFS vs DFS, how you parse the hostname, how you avoid cycles, and how you'd handle errors.

“I went with BFS mostly out of habit and then had to actually defend it, which I wasn't fully prepared for.”

View Post