AI Engineering Interview Questions — Practice Library | Velocode

300+ problems

Problem

Tags

Type

Semantic cache using embeddings for similarity matching

Reported at OpenAISWE (AI-Era)Medium

Access control filter placement before reranker causes leakage bug

Reported at CohereAI EngineeringMedium

Code Comprehension

Integration Drops Out Every Hour (Token Refresh)

Forward DeployedMedium

Code Comprehension

RAG Chatbot Ignoring Retrieved Documents, Using Training Data Instead

Reported at AnthropicAI EngineeringEasy

Code Comprehension

Retrieval deduplication: chunks from same document not filtered before ranking.

Reported at PineconeAI EngineeringMedium

Code Comprehension

Two Teams Want the Agent to Behave Differently

Forward DeployedHard

Hallucination Detection Pipeline Fails On Contradictory Claims Against Context

Reported at Scale AIAI EngineeringMedium

Code Comprehension

Token Budget Manager for Multi-turn LLM Applications

Reported at OpenAISWE (AI-Era)Easy

Build a Robust CSV Importer for Messy Customer Data

Forward DeployedEasy

Multi-tenant RAG: Secure Vector Index Isolation for Enterprise

Reported at OpenAIAI EngineeringHard

Template variable substitution with validation and defaults

Reported at CohereSWE (AI-Era)Easy

One Customer Can See Another Customer's Data

Forward DeployedHard

Code Comprehension

Conversation History Compression for Long-Running Agents

Reported at AnthropicSWE (AI-Era)Hard

Prompt Injection Vulnerability in Content Filter Implementation

Reported at AnthropicSWE (AI-Era)Medium

Code Comprehension

Agent Calls the Wrong Tool

Forward DeployedMedium

Code Comprehension

Fine-tuned model tokenizer mismatch with deployment environment configuration

Reported at MetaAI EngineeringHard

Code Comprehension

LLM Agent Tool Calls Exceed Context Limit Token Budget

Reported at Google DeepMindAI EngineeringMedium

Code Comprehension

Scheduled Reports Land on the Wrong Day

Forward DeployedMedium

Code Comprehension

Hierarchical cache breakpoints: system prompt, tool definitions, context windows, query batches

Reported at CursorAI EngineeringMedium

Sliding Window Rate Limiter for Concurrent LLM API Calls

Reported at DatabricksSWE (AI-Era)Medium

Duplicate Records After a Data Migration

Forward DeployedMedium

Code Comprehension

Trim conversation context to a token budget

Reported at GoogleSWE (AI-Era)Medium

Document chunking for vector databases with sentence-boundary preservation

Reported at PerplexitySWE (AI-Era)Easy

A Requested Feature Conflicts With Data Privacy

Forward DeployedHard

Loading more problems…