← AI Infrastructure System Design / 34

hardNew

Design an online and offline feature store without leaking future information into training

Trace event time and serving availability through a late-arriving feature. Build an executable historical join and keep online freshness separate from training correctness.

Updated Sep 2026 · Learn how AI infrastructure works through explanations, worked examples and diagrams, then practise applying it to interview questions.

Trace event time and serving availability through a late-arriving feature. Build an executable historical join and keep online freshness separate from training correctness.

The full answer is part of Premium. A free account includes more answers in each topic, but does not unlock this one.

20 answers per topic instead of 10, plus saved progress and bookmarks · no cardor unlock all 303 remaining answers · ₹2,000 / $25

The concepts behind this question

Use these concept explanations to review the ideas behind this question.

📐 AI Systems Design🔒 Premium

Model Metadata and Artifact StorageModel files and release records have different storage needs. Immutable artifacts and a controlled release pointer help prevent partially uploaded models from reaching users.

📐 AI Systems Design🔒 Premium

Evaluation and Data Pipeline InfrastructureData pipelines prepare traceable training inputs; evaluation pipelines compare model behavior. Both need versioned inputs, reproducible execution and useful failure reporting.

📐 AI Systems Design

Inference Platform ArchitectureAn inference platform connects API requests to model replicas. Learn how routing, batching, memory limits and scaling work together to meet a latency target.

📐 AI Systems Design🔒 Premium

Request Routing and Load Balancing for LLMsAn LLM router chooses a compatible replica, weighing queue time against reusable prompt state. Cached work helps only when the extra wait does not erase the saving.

Explore the full curriculum →

UP NEXT ON YOUR JOURNEY

Closest to this topicDesign the pipeline that produces 15 trillion training tokens: ingest, dedup, tokenize, shard, serve. Throughput per stage.Closest to this topicDesign a distillation pipeline that improves a target task without hiding capability regressions Closest to this topicTraining loss went flat at step 40k after descending normally. Walk me through how you debug it.

DISCUSSION · 0

No comments yet — be the first to share your approach.