Question Bank

Practice algorithm and coding problems to ace your technical interviews.

9 real interview questions collected so far

Questions I personally encountered in interviews — more coming as I interview! These are shown first with a red background.

Recommended Next

0 of 126 completed. Here's a great next challenge:

Easy

Valid Parentheses

stack, stringLeetCode #20

Your Progress

0 of 126 completed

Source

Difficulty

Status

Topic

Tag

Real

Have you built and maintained production-grade ML pipelines and implemented privacy/security controls for sensitive data?

Hard

mlops, security, privacy +5

How is AutoGen created behind the scene? Can you create a talking multi-agent system without AutoGen?

Hard

autogen, multi-agent, conversation-loop +6

How do you make sure high traffic is possible in FastAPI? How do you make it scalable?

Hard

fastapi, scalability, horizontal-scaling +6

What is overfitting? How do you prevent it?

Easy

fundamental, regularization, must-know +1

What is Google ADK (Agent Development Kit)? How does it compare to other agent frameworks?

Hard

google-adk, ai-agents, gemini +4

How do you integrate with external APIs in Python and Java? Compare the approaches and best practices.

Medium

api-integration, python, java +4

Can you walk us through a recent AI or GenAI project you've worked on?

Medium

genai, llm, project-experience +3

Tell me about a time you had to design an API or service. How did you approach it and what decisions did you make around scalability and maintainability?

Hard

api-design, system-design, scalability +4

Explain LoRA and QLoRA. How do they work and when would you use them?

Hard

lora, qlora, fine-tuning +4

Have you used Google BigQuery? How do you optimize queries and manage costs in BigQuery?

Medium

bigquery, google-cloud, data-warehouse +4

When do you use Design Patterns in your programming? Which ones are your 'go-to' or favorite patterns?

Medium

design-patterns, software-engineering, architecture +6

What are the differences between FastAPI and Flask? When would you choose one over the other?

Medium

fastapi, flask, python +5

What do you know about memory management in AI agents? What is LangMem?

Hard

langmem, langgraph, agent-memory +6

How do you deliver a multi-agent system to production?

Hard

multi-agent, deployment, production +7

How many types of caches are there in a system?

Medium

caching, system-design, redis +5

How do you handle high traffic in FastAPI?

Hard

fastapi, scalability, high-traffic +7

How do you think about integrating AI capabilities into existing products or systems?

Hard

ai-integration, strategy, product-thinking +4

Explain the bias-variance tradeoff.

Easy

fundamental, theory, must-know +1

Tell me about your last ML project. What challenges did you face and how did you solve them?

Medium

langgraph, multi-agent, optimization +3

Explain backpropagation. How does it work?

Medium

deep-learning, optimization, must-know +1

What's the difference between Tools and MCP (Model Context Protocol)?

Medium

mcp, tools, llm-integration +3

Explain precision, recall, and F1 score. When to optimize for which?

Easy

evaluation, metrics, must-know +1

Have you implemented an MCP server? Walk me through how you would build one.

Hard

mcp, fastmcp, server +3

What is the attention mechanism in transformers?

Medium

nlp, transformers, deep-learning +2

What is the A2A (Agent-to-Agent) protocol? How does it compare to MCP?

Hard

a2a, google, multi-agent +3

Tell me about your LangGraph experience. How do you ensure correct results and visualize agent connections?

Hard

langgraph, multi-agent, validation +3

Compare zero-shot, few-shot, and chain-of-thought prompting. When would you use each?

Easy

llm-engineering, prompting, must-know +1

When would you fine-tune an LLM vs using prompting?

Medium

llm-engineering, fine-tuning, must-know +1

How do you deploy ML models in production? Walk me through your approach.

Hard

deployment, fastapi, docker +3

What model parameters do you adjust when using LLM APIs? What API types are available?

Medium

openai, api, parameters +3

Explain function calling in LLMs. How does the model decide when to use tools?

Medium

llm-engineering, function-calling, tools +1

What causes LLM hallucinations? How do you detect and prevent them?

Medium

llm-engineering, hallucination, must-know +1

Compare GPT-4o vs GPT-4o-mini vs other models. When do you use each?

Medium

model-selection, gpt-4o, cost-optimization +2

What vector databases have you used? Explain the different similarity metrics.

Medium

vector-db, pinecone, chromadb +4

How do you evaluate LLM outputs? What metrics matter?

Medium

llm-engineering, evaluation, metrics +1

How do you manage context windows in LLM applications?

Medium

llm-engineering, context-window, optimization +1

You don't have customer data yet. How do you build and validate an ML system?

Hard

data-strategy, synthetic-data, few-shot +3

How do you evaluate RAG system accuracy? Have you used RAGAS?

Hard

ragas, evaluation, rag +4

Explain RAG (Retrieval-Augmented Generation). When and how would you use it?

Medium

llm-engineering, rag, retrieval +2

How do you evaluate LLM outputs? What metrics and methods do you use?

Medium

llm-evaluation, bleu, rouge +3

Compare GPT-4, Claude, and open-source LLMs. When would you use each?

Medium

llm-engineering, model-selection, cost-optimization +1

How do you detect and prevent hallucinations in LLM applications?

Hard

hallucination, groundedness, factuality +3

How do you ensure LLMs output valid structured data (JSON, specific formats)?

Medium

llm-engineering, structured-outputs, validation +1

How do you run A/B tests for LLM-powered features?

Hard

a/b-testing, experimentation, production-ml +2

AI Orchestration: How do you coordinate multiple AI agents in workflows?

Hard

llm-engineering, multi-agent, orchestration +2

Explain PCA (Principal Component Analysis). When and how do you use it?

Medium

dimensionality-reduction, feature-engineering, unsupervised +1

How do you evaluate and test an LLM-based system before deploying to production? What metrics do you track?

Hard

llm-testing, production-ml, rag-evaluation +4

How do you handle model versioning and rollback in production? What happens if a new model performs worse than expected?

Hard

model-versioning, rollback, mlops +4

How do you choose the number of epochs? Explain early stopping and gradient descent optimizers.

Medium

training, optimization, epochs +2

When would you use fine-tuning vs RAG vs prompt engineering? How do you decide?

Medium

llm-engineering, fine-tuning, rag +3

Walk me through building a production fine-tuning pipeline from start to finish. What are the key steps?

Hard

fine-tuning, production-ml, mlops +6

How do you prepare training data for fine-tuning? How much data do you need and how do you ensure quality?

Medium

llm-engineering, fine-tuning, data-preparation +2

What is a knowledge graph and how does it differ from a traditional relational database?

Medium

knowledge-graph, neo4j, graph-database +5

What is an ontology in the context of Knowledge Graphs? Why is it important?

Medium

knowledge-graph, ontology, rdf +5

How can Knowledge Graphs help reduce hallucinations in LLM applications?

Hard

knowledge-graph, hallucination, rag +5

Can you explain what MCP (Model Context Protocol) is and why Anthropic created it? What problem does it solve?

Medium

mcp, model-context-protocol, anthropic +5

Briefly describe an AI-powered system or feature you've shipped to production. What problem did it solve, and what was your role?

Medium

ai-production, project-experience, behavioral +4

When would you choose A2A over LangGraph?

Hard

a2a, langgraph, multi-agent +4

What performance/load testing tools have you used? How do you create realistic load tests?

Medium

load-testing, performance, locust +3

What do you do when traffic is too high? How do you handle traffic spikes?

Medium

scaling, auto-scaling, rate-limiting +3

How do you identify and fix slow endpoints?

Medium

performance, optimization, profiling +4

What challenges have you faced in full-stack development and deployment? How did you solve them?

Medium

deployment, full-stack, docker +5

Describe a technical tradeoff you had to make when building or scaling a system (AI-related or otherwise). What were the constraints, and how did you decide?

Hard

system-design, tradeoffs, architecture +4

How do you approach feature engineering for ML models? Walk me through your process.

Medium

feature-engineering, data-science, preprocessing +2

How do you ensure data quality in production ML pipelines? What tools and practices do you use?

Hard

data-quality, great-expectations, pandera +3

Explain your approach to data aggregation for analytics and ML features. How do you handle different time windows?

Medium

aggregation, sql, window-functions +2

How do you validate data at different stages of a pipeline? What tools do you use?

Medium

data-validation, pandera, great-expectations +2

How do you optimize slow SQL queries for large datasets? Walk me through your debugging process.

Hard

sql, optimization, indexing +2

Compare ETL vs ELT approaches. When would you use each?

Medium

etl, elt, dbt +3

How do you handle missing data in ML pipelines? What are the different strategies?

Medium

missing-data, imputation, preprocessing +2

How do you detect and handle data drift in production ML systems?

Hard

data-drift, monitoring, mlops +2

You mention 'Python (Advanced)' on your resume. Explain Python's GIL and how it affects multi-threading in ML workloads.

Hard

python, concurrency, performance +2

Explain Python's type hints and how you use them in production ML code. Why are they important?

Medium

python, type-hints, code-quality +2

Walk me through your code structure for a production ML API. What design patterns do you use?

Hard

python, architecture, design-patterns +2

How does async work in Python and Java? Compare their async models.

Hard

python, java, async +3

How does concurrency work in Python and Java? Explain the key differences.

Hard

python, java, concurrency +4

How do you implement a production service in Python and Java? Walk me through the key components.

Hard

python, java, fastapi +4

Your ML interview website is built with TypeScript. Why TypeScript over JavaScript for this project?

Medium

typescript, frontend, type-safety +2

You mention React experience from Android/React Native. How did you apply that to building web interfaces for your ML projects?

Medium

react, frontend, gradio +2

Why did you choose Next.js for your ML interview website instead of plain React?

Medium

next.js, ssg, ssr +3

You've built 4 production FastAPI services. What are your REST API design principles?

Hard

rest-api, fastapi, api-design +2

You use pgvector for semantic search in your RAG chatbot. How does it work and why PostgreSQL over a dedicated vector DB?

Hard

postgresql, pgvector, vector-search +2

You mention 35% cache hit rate with Redis. Walk me through your caching strategy.

Medium

redis, caching, optimization +2

You fine-tuned BERT for sentiment analysis (89% F1). Explain the fine-tuning process step-by-step.

Hard

bert, nlp, fine-tuning +2

You used ResNet50 for image classification (94% accuracy). Why ResNet over other architectures?

Medium

resnet, computer-vision, transfer-learning +1

You implemented Grad-CAM for model interpretability. How does it work and why is it useful?

Hard

grad-cam, interpretability, explainability +2

You mention AWS (S3, EC2, SageMaker). How have you used these services in your ML projects?

Medium

aws, s3, ec2 +3

Implement K-Means Clustering

Medium

clustering, unsupervised, classic-ml +2

Implement Linear Regression with Gradient Descent

Medium

regression, gradient-descent, optimization +2

Implement Softmax and Cross-Entropy Loss

Easy

neural-networks, classification, loss-functions +2

Golden Brick Road

Medium

bfs, graph, shortest-path +3

Fibonacci Number

Easy

dynamic-programming, recursion, easy +2

Valid Palindrome

Easy

two-pointers, string, easy +2

Reverse an Array

Easy

two-pointers, array, easy +2

Question Bank

Recommended Next

Valid Parentheses

Linear Regression Implementation

PyTorch Tensor Operations

Python OpenAI LangChain RAG

PyTorch Statistics

Python OpenAI Haystack RAG Project

Rate Limiter Implementation

Filter Words by Letters

Factorize String Extremities

Sequence Join Point

Valid Parentheses

Valid Anagram

Contains Duplicate

Number of Islands

Invert Binary Tree

Merge Two Sorted Lists

Reverse Linked List

Coin Change

Linked List Cycle

Maximum Depth of Binary Tree

Two Sum

Same Tree

Merge K Sorted Lists

Top K Frequent Elements

Group Anagrams

Climbing Stairs

Clone Graph

Kth Largest Element in an Array

Min Stack

Binary Tree Level Order Traversal

System Design: AI Customer Support with AWS Bedrock

System Design: LangGraph Log Analysis & Slack Notification System

System Design: MLOps Pipeline with MLflow & Kubeflow

Have you built and maintained production-grade ML pipelines and implemented privacy/security controls for sensitive data?

How is AutoGen created behind the scene? Can you create a talking multi-agent system without AutoGen?

How do you make sure high traffic is possible in FastAPI? How do you make it scalable?

What is overfitting? How do you prevent it?

What is Google ADK (Agent Development Kit)? How does it compare to other agent frameworks?

How do you integrate with external APIs in Python and Java? Compare the approaches and best practices.

Can you walk us through a recent AI or GenAI project you've worked on?

Tell me about a time you had to design an API or service. How did you approach it and what decisions did you make around scalability and maintainability?

Explain LoRA and QLoRA. How do they work and when would you use them?

Have you used Google BigQuery? How do you optimize queries and manage costs in BigQuery?

When do you use Design Patterns in your programming? Which ones are your 'go-to' or favorite patterns?

What are the differences between FastAPI and Flask? When would you choose one over the other?

What do you know about memory management in AI agents? What is LangMem?

How do you deliver a multi-agent system to production?

How many types of caches are there in a system?

How do you handle high traffic in FastAPI?

How do you think about integrating AI capabilities into existing products or systems?

Explain the bias-variance tradeoff.

Tell me about your last ML project. What challenges did you face and how did you solve them?

Explain backpropagation. How does it work?

What's the difference between Tools and MCP (Model Context Protocol)?

Explain precision, recall, and F1 score. When to optimize for which?

Have you implemented an MCP server? Walk me through how you would build one.

What is the attention mechanism in transformers?

What is the A2A (Agent-to-Agent) protocol? How does it compare to MCP?

Tell me about your LangGraph experience. How do you ensure correct results and visualize agent connections?

Compare zero-shot, few-shot, and chain-of-thought prompting. When would you use each?

When would you fine-tune an LLM vs using prompting?

How do you deploy ML models in production? Walk me through your approach.

What model parameters do you adjust when using LLM APIs? What API types are available?

Explain function calling in LLMs. How does the model decide when to use tools?

What causes LLM hallucinations? How do you detect and prevent them?

Compare GPT-4o vs GPT-4o-mini vs other models. When do you use each?

What vector databases have you used? Explain the different similarity metrics.

How do you evaluate LLM outputs? What metrics matter?

How do you manage context windows in LLM applications?

You don't have customer data yet. How do you build and validate an ML system?

How do you evaluate RAG system accuracy? Have you used RAGAS?

Explain RAG (Retrieval-Augmented Generation). When and how would you use it?

How do you evaluate LLM outputs? What metrics and methods do you use?

Compare GPT-4, Claude, and open-source LLMs. When would you use each?

How do you detect and prevent hallucinations in LLM applications?

How do you ensure LLMs output valid structured data (JSON, specific formats)?

How do you run A/B tests for LLM-powered features?

AI Orchestration: How do you coordinate multiple AI agents in workflows?