Event held August 27. 2026

Problem solving interviews are often a very important interview round in the Yelp technical hiring process. They test how you approach real-world scenarios, talk through your reasoning, and write working code under time pressure. This Cafecito with Yelp picked up where Leveling Up Your Interview Skills Part 1 left off. In April, we covered technical communication and behavioral interviews. This time, the event focused on problem solving interviews: what actually happens, how interviewers evaluate you, and what creates a strong interview. Read a summary from each part of the event below.

What to Expect: Problem Solving at Yelp, From Interviewing to Onboarding

At Yelp, the problem solving interviews sit within the panel stage: four back-to-back one-on-one sessions, each 45 minutes. Emily K., a senior technical recruiter, broke down what to expect. You're given a real-world business scenario, assess the problem collaboratively with your interviewer, and implement a solution live. Before the panel interview stage, candidates typically have gone through the Technical Communication Behavioral round and a coding challenge (depending on the role), for which we provide practice tests so candidates can better prepare.

Julio C.B.O., a software engineer based in Tijuana, shared what it's actually like once you're hired. From day one, he was paired with a mentor who helped him navigate the tech stack and codebase. The onboarding is structured with weekly ramp-up materials, and he had his first code merged within two weeks. Even fully remote, his team stays connected through quarterly social events ranging from cooking classes to Lego making sessions.

Where Does AI Fit In?

During Yelp’s problem solving interviews, GenAI is allowed as a supplement, but it should never replace your own thinking. You can use HackerRank's built-in AI assistant for help with syntax checks, clarifying questions, or verifying an approach you're considering. What you can't do is use any third-party AI tools or external note-taking apps, generate complete solutions, or paste the entire interview prompt in for a direct answer. Your evaluation criteria are the same whether you use AI or not. If you do use the HackerRank assistant, just let your interviewer know how you're using it and always explain your own thinking alongside it.

What Does Yelp Look For in a Problem Solving Interview?

Blake L.E., a tech lead, and Cynthia M., a machine learning engineer, ran two live mock interviews using the same problem: classify user input text like "fix a leaky faucet" into service categories like "plumbing" or "electrical" based on keyword matching. They swapped roles between scenarios, and it was clear which went well and which didn't. Even the audience pointed out the differences with ease.

In the first scenario, the candidate stayed silent while thinking, jumped into code without asking questions, ignored hints, and gave vague answers about complexity. In the second, the candidate restated the problem first, broke it into smaller pieces, asked clarifying questions throughout, and responded to feedback right away. Here are some tips and advice to keep in mind:
Before you code

Restate the problem in your own words. Clarifying your understanding of the problem is half the battle. Repeat it back before you start coding.
Ask clarifying questions and respond to feedback. Confirm assumptions, ask about edge cases, and when your interviewer gives a hint, take it. Treat them as a collaborator, not an evaluator.
Start simple, then build up. Solve a smaller piece first, then layer on complexity. Don't try to tackle the full problem in one go.
Consider a framework like REACTO (Repeat, Examples, Approach, Code, Test, Optimize) to keep yourself organized under pressure.

While you code

Think out loud. Silence makes it hard for your interviewer to help. If they can follow your thinking, they can steer you before you go down the wrong path.
Write clean code with clear names. Use meaningful variable and function names. It makes your code easier to follow and shows you care about readability.
Talk through time and space complexity as you go. Don't wait to be asked. Verbalize why you chose a certain approach and where it could be more efficient.
Pause before switching approaches. If you're about to make a big change or try something different, confirm with your interviewer first instead of silently changing direction.

Remember: your interviewer is a resource

Ask your interviewer for syntax help. You can use them as a resource. They've seen the question many times and don't expect you to memorize every API.
If you're stuck, say so. Your interviewer can reframe the problem or pivot to a different part entirely, but only if you tell them where you're blocked.

"We want to gather data points that you know how to do the job," Blake said. "We are very much on your side here."

Resources to Help You Prepare [LINK]

We put together a resources page with everything mentioned during the event and more. Blake specifically called out Advent of Code as a good starting point for general practice before you apply, noting the difficulty is close to Yelp's first-pass problems. The page also includes Yelp's interview prep materials, practice coding challenges, and links to additional tools like Cracking the Coding Interview and LeetCode's data structures and algorithms course.

Watch the Full Session

This blog covers the key takeaways, but the full recording has the complete mock interviews, audience Q&A, and all the details we couldn't fit here. Whether you're actively preparing for a Yelp interview or just starting to think about your next role, this is one to keep in your back pocket.

¡Éxito! 🎉

About the Speakers

Want to find out about our next tech event? Sign up to be notified and learn about upcoming Yelp recruiting, networking, and professional events.