Meta·Machine Learning Engineer·Technical Phone Screen
Jun 2026
Fundamentals round at Meta for an MLE role covering optimizers, neural scaling laws, and clustering. Pretty broad for a single session but the questions were genuinely deep, not just name-dropping methods.
- Walk through the progression from plain SGD to Adam. What does each step add, and why is Adam the default for training transformers?
- What is the actual difference between L2 regularization added to the loss and decoupled weight decay as in AdamW?
- How does learning rate scheduling with warmup and decay fit into optimizer design, and why is warmup particularly important?
- State the power-law form of neural scaling laws and explain what the compute-optimal result says about splitting a fixed compute budget between model size and training tokens.
- If you have a fixed compute budget but a hard inference latency constraint, how does that change which model size you'd actually ship?
- Compare k-means and GMM fitted by EM: what objective does each optimize, what's the difference in how they assign points to clusters, and what's the formal relationship between the two algorithms?
- Your GMM's EM is diverging because covariance matrices are collapsing toward zero. What's going wrong and how do you fix it?
- How would you choose the number of clusters k for k-means versus the number of components for a GMM, and do the model selection tools differ?
- Why does Adam sometimes generalize worse than well-tuned SGD with momentum, and when does that gap actually matter in practice?
“I traced the chain fine: gradient descent, then momentum to smooth updates, then RMSProp to scale per-parameter, then Adam combining both with bias correction.” The rest of the author's notes on Machine Learning Engineer interview at Meta, Technical Phone Screen round, covers how they worked through the question, what the panel pushed back on, and what they would do differently.
View Post