Luma AI interview questions
AI startup · 3 questions reported
- junior 0
- mid 2
- senior 1
These are 3 technical questions candidates reported being asked in Luma AI interviews for AI engineer, ML engineer and applied AI engineer roles, each with a model answer you can say out loud in a minute or two. They cover regression tests, training and multimodal: 0 junior, 2 mid-level and 1 senior questions. Every question comes from a public interview write-up that names Luma AI.Updated
Luma AI prompt versions and regression tests questions
Model answer
Treat prompts as versioned config: a registry stores each prompt's template, model, parameters and output schema as an immutable version that the app references by ID. Every change runs a regression suite of golden cases and judge checks before promotion.
Scroll sideways to see the whole diagram
Anything that can change the output counts as a change, not only edits to the prompt text.
Text version
- Anything that can change the output counts as a change, not only edits to the prompt text.
- Run the whole fixed suite automatically, the way unit tests run on code.
- Gate on broken cases, not only on the total: a better average can still hide a broken case you care about.
- When something slips through to users, add it as a case so it can't silently return.
Roll out like a feature flag, with instant rollback. Log the prompt version and model with every output, so production quality can be sliced by version.
Luma AI training and optimisation questions
Model answer
Start with the data, since a distribution mismatch splits the curves whatever the model does. Check train and test share normalisation, scale and noise level. Then the loop: the loss compares the output with the clean target, not the noisy input; zero_grad, backward and step run in order; train and eval modes are set. Overfit one batch first: if loss won't fall there, the bug is in the code.
Luma AI multimodal models questions
Model answer
Two paths over one model fleet. Streaming: normalise the text, then generate and vocode audio in small chunks so playback starts fast; these requests get priority on warm GPUs. Batch: split long text at sentence boundaries, synthesise in parallel on spare capacity, stitch with consistent prosody, and return through a job API and storage.
Quality: human listening ratings plus automatic checks, such as word error rate from speech recognition and speaker similarity. Voice authorization: clone a voice only with the speaker's recorded consent, bind voices to accounts, and watermark output.
Each question comes from a public interview write-up or video that names Luma AI; the answers are ours. Not affiliated with Luma AI.