Zenaique
Topics
Practice
Study
Browse
Reference
Pricing
Search…
⌘K
Formats
/
Multiple Choice
Multiple Choice
617 questions
Topics
Difficulty
Role
Experience
Companies
Sort
Newest
Quality
Difficulty ↑
Difficulty ↓
Questions
You upsert a document and query for it 200 milliseconds later, but it never appears. Pick the most likely cause.
Multiple Choice
Easy
Premium
A news product ingests…
Multiple Choice
Medium
Premium
Pick between int8 scalar…
Multiple Choice
Medium
Pick the right billing model for a search feature that bursts to 2,000 QPS one hour per day.
Multiple Choice
Easy
One billion vectors must live mostly on cheap storage with a 200 ms budget. Decide between SPANN style and DiskANN style designs.
Multiple Choice
Hard
During a background reindex, p99 latency triples while p50 barely moves. Pick the explanation.
Multiple Choice
Medium
Your nightly index rebuild takes 14 hours on CPU and the maintenance window is 2 hours. Pick the most direct fix.
Multiple Choice
Easy
Pick the workload where an object storage backed serverless vector database beats a RAM resident cluster on cost.
Multiple Choice
Medium
EU users see yesterday's documents while US users see today's. Debug this geo replicated vector deployment.
Multiple Choice
Medium
Premium
A tenant demands full…
Multiple Choice
Medium
A filtered query with k=20 returns only 4 hits even though 50,000 vectors pass the filter. Pick the fix.
Multiple Choice
Medium
Your filters hit anywhere from 0.5% to 50% of the corpus depending on the query. Pick the filtered ANN strategy that survives both extremes.
Multiple Choice
Hard
Estimate the parameters weight tying saves a 128k-vocab model at d_model 2048
Multiple Choice
Easy
Premium
Pinpoint the mechanism that…
Multiple Choice
Medium
Reason about KV cache growth in a sliding window model pushed to 32k tokens
Multiple Choice
Medium
Explain why raising the RoPE base theta helps a model accept longer contexts
Multiple Choice
Medium
Pick the real reason Llama class models swapped LayerNorm for RMSNorm
Multiple Choice
Easy
Premium
Explain how a layer…
Multiple Choice
Medium
Choose the positional scheme that fails hardest when context quadruples at inference
Multiple Choice
Medium
Identify what changes when a block computes attention and FFN in parallel, PaLM style
Multiple Choice
Medium
Interpret what the logit lens reveals when you unembed intermediate layers
Multiple Choice
Hard
Identify which tensors your inference server actually stores in the KV cache
Multiple Choice
Easy
Justify why hybrid SSM models like Jamba still keep a slice of attention layers
Multiple Choice
Medium
Anticipate what breaks when you bolt naive early exit onto a decoder at inference
Multiple Choice
Medium
Annotators keep marking ties between responses: decide how to treat them
Multiple Choice
Easy
Showing 1–25 of 617
← Prev
Next →
Airbnb
Anduril
Goldman Sachs
Jasper
Anyscale
Persistent
Comet Ml
Graphcore
Airbnb
Flowise
Arize Ai
Datadog
Doordash
Patronus
Fireworks Ai
Kpmg
Databricks
Humanloop
Kpmg
Patronus
Comet Ml
Deloitte
Dify
Sarvam
Krutrim
Mphasis
Persistent
Polyai
Baseten
Jump Trading
Evenup
Infosys
Elevenlabs
Hcl
Ai4bharat
Neptune Ai
Induced Ai
Perplexity
Anyscale
Goldman Sachs
Banana Dev
Canva
Coreweave
Harvey
Jasper
Linkedin
Palantir
Perplexity
Meesho
Pinecone