performance
FDE interview questions tagged performance, across every topic.
10 questions · 1 unlocked for you
Concepts behind "performance"
The curriculum that explains the ideas these questions test.
Foundational
Big-O That Actually MattersOn a deployment, Big-O is not a whiteboard puzzle; it is the one calculation that tells you whether the customer's data fits in the approach you picked. The skill is spotting the term that dominates at their scale, knowing when brute force dies and you need an index or ANN, and recognizing when constant factors and memory decide the outcome instead of the exponent.💻 Coding & Engineering Craft
Core
Latency OptimizationMeasure p50, p95, and p99 before you touch anything, then find where the time actually goes: tokenization, retrieval, inference, or post-processing. A naive RAG pipeline that takes 1.5 seconds can usually reach sub-100ms perceived latency by caching, parallelizing retrieval, picking a smaller model, and streaming the first token, in that order of payoff.⚙️ System Design for AI in ProductionSign in
Core
GPU Architecture and ExecutionA GPU is not a fast CPU. It runs thousands of threads in lockstep groups called warps across many streaming multiprocessors, under the SIMT model, and its real constraint is moving data through a memory hierarchy that spans fast on-chip registers and shared memory down to slow off-chip HBM. Understanding occupancy, coalesced memory access, and warp divergence is what separates a kernel that hits peak throughput from one that leaves 90% of the chip idle.🖥️ ML Infrastructure & ServingSign in
Advanced
Query Optimization and Execution PlansWhen a customer says a query is slow, the answer lives in the execution plan, not in intuition. Reading a plan tells you whether the engine is scanning when it should seek, exploding rows in a join, or shuffling across a network, and those three causes have nothing in common except the symptom.🗄️ Data & SQL Engineering🔒 Premium
Advanced
Spark Internals and Performance TuningSpark is fast until a shuffle, and almost every Spark performance problem is a shuffle problem wearing a disguise. Knowing which transformations force data across the network, and how to spot skew in the stage view, is the difference between a job that finishes in four minutes and the same job running for an hour.🗄️ Data & SQL Engineering🔒 Premium