AI Engineering Interview Questions — Practice Library

Start here · SWE (AI-Era)

Start here: Software Engineer (AI-Era)

Six problems, easy to hard. Work through them in order — each one builds on the last.

0/ 6 done

1
Chatbot Ignores Prompt Instructions Despite Correct Configuration Setup
Easy
Start →
2
Vector Embedding Space Mismatch Between Query And Document Processing
Easy
Start →
3
RAG Chatbot Fails to Answer Despite Successfully Retrieving Relevant Documents
EasyGoogle
Pro
4
Cache overhead exceeds benefits due to poor invalidation strategy and key design
Medium
Start →
5
Reranker degrades pipeline performance despite correct isolation testing behavior.
MediumCohere
Pro
6
Vector Store Dimension Mismatch: Ingestion and Retrieval Pipeline Synchronization
HardPinecone
Pro

37 problems

Problem

Tags

Type

Document chunking for vector databases with sentence-boundary preservation

Reported at PerplexitySWE (AI-Era)Easy

Coding

Semantic cache using embeddings for similarity matching

Reported at OpenAISWE (AI-Era)Medium

Coding

Stateless Chatbot Losing Conversation History Between Turns

SWE (AI-Era)Easy

Code Comprehension

Agent Function Calling JSON Generation Produces Malformed Output Intermittently

Reported at OpenAISWE (AI-Era)Medium

Code Comprehension

Template variable substitution with validation and defaults

Reported at CohereSWE (AI-Era)Easy

Coding

Cache overhead exceeds benefits due to poor invalidation strategy and key design

SWE (AI-Era)Medium

Code Comprehension

LLM output validation with schema enforcement and type checking

Reported at MetaSWE (AI-Era)Medium

Coding

Reranker degrades pipeline performance despite correct isolation testing behavior.

Reported at CohereSWE (AI-Era)Medium

Code Comprehension

Chatbot Ignores Prompt Instructions Despite Correct Configuration Setup

SWE (AI-Era)Easy

Code Comprehension

Silent data drift detection in production ML pipeline accuracy collapse

Reported at DatabricksSWE (AI-Era)Medium

Code Comprehension

Stream Aggregator Batches LLM Tokens Into Complete Sentences

Reported at GoogleSWE (AI-Era)Medium

Coding

LLM Summarizer Producing Hallucinated Content At High Failure Rate

SWE (AI-Era)Medium

Code Comprehension

RAG evaluation metrics masking real-world answer quality failures

Reported at AnthropicSWE (AI-Era)Medium

Code Comprehension

Async batch processor silent failures with missing error handling logic

Reported at MetaSWE (AI-Era)Medium

Code Comprehension

LLM Summarization Tool Cuts Off Mid-Sentence Completion Issue

SWE (AI-Era)Easy

Code Comprehension

LLM Cache Bug: Same Response Returned Regardless Model Parameter

Reported at MetaSWE (AI-Era)Easy

Code Comprehension

Token Budget Manager for Multi-turn LLM Applications

Reported at OpenAISWE (AI-Era)Easy

Coding

Vector Embedding Space Mismatch Between Query And Document Processing

SWE (AI-Era)Easy

Code Comprehension

Resilient LLM API calls with exponential backoff retry logic

Reported at AnthropicSWE (AI-Era)Easy

Coding

Trim conversation context to a token budget

Reported at GoogleSWE (AI-Era)Medium

Coding

Semantic Search Returns Identical Results Regardless Input Query

SWE (AI-Era)Easy

Code Comprehension

AI Agent Prompt Injection Attack Via Support Ticket Content

Reported at AnthropicSWE (AI-Era)Hard

Code Comprehension

Token Counter Mismatch Between Dashboard Estimates And Actual LLM Costs

Reported at OpenAISWE (AI-Era)Easy

Code Comprehension

Agent infinite loops on tool due to missing success detection logic

SWE (AI-Era)Medium

Code Comprehension

Loading more problems…