Glassdoor users rated their interview experience at Luma AI as 26.7% positive with a difficulty rating score of 3.2 out of 5 (where 5 is the highest level of difficulty). Candidates interviewing for Staff Software Engineer and Research Engineer II rated their interviews as the hardest, whereas interviews for Software Engineer and Recruiter roles were rated as the easiest.
The hiring process at Luma AI takes an average of 25 days when considering 15 user submitted interviews across all job titles. To compare, the average duration of hiring at similar companies like BlackRock, Inc. is 14 days, Fabricated Software, Inc. is 2 days, and Apple Inc. is 21 days. Candidates applying for Research Scientist had the quickest hiring process (on average 21 days), whereas Chief of Staff roles had the slowest hiring process (on average 30 days).
They have really quick communication and fast interview process, usually will get back within 2-3 days. You will have opportunity to select days and times of interviews. Pretty open to rescheduling as well.
I applied online. I interviewed at Luma AI (Palo Alto, CA)
Interview
It was a take home test that asked to build something from end to end. It takes like 8 to 12 hours. Not only they seek for a MVP, but a scalable API with public hosting! In other words, they want you build something ready for market! It is one day of entirely unpaid work, which I felt is unrespectful with our time. I have got it done, but I do not want to submit it
Interview questions [1]
Question 1
Choose a Problem
Pick the one that excites you most. These are deliberately open-ended — we want to see what directions you take and what decisions you make.
1. Reverse-Engineer an Undocumented API
Pick any website that doesn't have a public API. Reverse-engineer how it works under the hood, then build a useful tool on top of it.
Figure out the API, then build something real with it.
2. Fix Something Annoying
Pick a website you use daily. Identify something that genuinely annoys you about it. Build a browser extension that fixes it.
Find the annoyance, ship the fix.
3. Clone and Improve
Pick one specific feature or interaction from an app you admire. Rebuild it, then make it better — faster, cleaner, more thoughtful, whatever "better" means to you.
Rebuild it, improve it, and explain what you changed and why.
I applied through a recruiter. I interviewed at Luma AI in May 2026
Interview
I interviewed for a Staff-level engineering role at Luma AI.
The recruiter screen was positive and I was moved to a take-home assessment. The assignment was not a small coding exercise. It required building a complete end-to-end product, making it deployable/publicly accessible, writing documentation, recording a video walkthrough, and submitting AI session history. It was estimated at 8–12 hours, which is essentially a full unpaid working day if done seriously.
The technical challenge itself was interesting, but the communication around the process was poor. Role alignment was unclear, follow-up was slow, and after submitting the assignment I received a generic rejection with no specific feedback. I then followed up politely asking for brief feedback, especially given the size of the assignment, and received no response.
The overall experience felt one-sided: Luma asks candidates for a significant amount of unpaid work, but does not provide meaningful communication or feedback in return. Candidates should understand that they may spend a full day building, deploying, documenting, and recording a project, then receive only a template rejection or no useful follow-up.
Interview questions [1]
Question 1
A full-day take-home assignment using AI coding tools, including a deployable end-to-end product, documentation, video walkthrough, and AI session history.