Overview
The AI and GenAI engineer interview, answered, pitched for candidates with up to five years of experience. Fifteen modules run from AI fundamentals (statistics, machine learning, deep learning and NLP) through transformers, LLMs, prompt engineering, fine-tuning, memory and databases, and backend work with FastAPI. They continue with GenAI system design, agents, RAG, evaluation and testing, serving, latency and cost, multimodal and voice AI, safety and security, and practical build and project rounds. Every module ends with hands-on and scenario questions, and many answers carry a diagram.
What you will learn
- Answer the math, ML and deep learning fundamentals interviewers still ask
- Explain transformers and LLMs from attention to decoding
- Apply prompt engineering and choose when to fine-tune
- Build GenAI backends with FastAPI and deploy them
- Design RAG and agent systems and evaluate them
- Reason about serving, latency and cost
- Handle multimodal, voice, safety and security questions
- Talk through real builds and projects in practical rounds
Included with the kit
- 108 written parts, yours for good
- 28h 45m of reading, measured not estimated
- Written for the intermediate level
- Every future revision included
- UPI, cards and netbanking
Prerequisites
- Python and basic machine learning
- Some experience calling LLM APIs
Curriculum
15 sections · 108 parts · 28h 45m01Module 1: AI Fundamentals5 parts · 167 min
Build the math, machine learning, and deep learning foundations that every GenAI system rests on.
- Statistics & Mathematics for AI43 min
- Machine Learning Fundamentals46 min
- Deep Learning & NLP Foundations51 min
- AI Fundamentals: hands-on15 min
- AI Fundamentals: scenarios12 min
02Module 2: Transformers8 parts · 104 min
Understand how transformers work inside, from attention and positional encoding to training objectives and MoE.
- Core Architecture & Model Types18 min
- Attention Mechanisms7 min
- Positional Encoding & Long Context10 min
- Pooling & Training Objectives13 min
- Mixture of Experts & Tokenization14 min
- Transformers for Practitioners14 min
- Transformers: hands-on17 min
- Transformers: scenarios11 min
03Module 3: LLMs7 parts · 110 min
Control LLM behavior, tokens, context, and cost, and handle real production pressure scenarios.
- LLM Deep Questions38 min
- Scenario-Based LLM Questions14 min
- Reasoning Models10 min
- Model Selection10 min
- Modern LLM API Features10 min
- LLMs: hands-on18 min
- LLMs: scenarios10 min
04Module 4: Prompt Engineering10 parts · 104 min
Design reliable prompts for standalone tasks, RAG pipelines, and agent systems.
- Core Prompting Concepts10 min
- Prompt Structure & Design11 min
- Prompting for RAG & Agents6 min
- Production Prompt Management6 min
- Context Engineering10 min
- Structured Outputs & Tool Prompts10 min
- Prompting Reasoning & Multimodal Models10 min
- Prompt Testing & Optimization10 min
- Prompt Engineering: hands-on18 min
- Prompt Engineering: scenarios13 min
05Module 5: Fine-Tuning & Model Adaptation10 parts · 110 min
Decide when and how to adapt models with PEFT, LoRA, data strategy, and alignment methods.
- Core Fine-Tuning Concepts18 min
- PEFT, LoRA & Efficient Methods10 min
- Data & Training Strategy8 min
- Alignment & RL-Based Methods8 min
- Production & Decision-Making6 min
- Preference Tuning in Practice9 min
- Data for Fine-Tuning11 min
- Serving & Evaluating Fine-Tunes9 min
- Fine-Tuning & Model Adaptation: hands-on17 min
- Fine-Tuning & Model Adaptation: scenarios14 min
06Module 6: Memory, Databases & Caching7 parts · 78 min
Design memory, storage, vector infrastructure, and caching layers for GenAI and agent systems.
- Memory Design in AI Systems8 min
- Databases for AI Systems9 min
- Vector Databases & Retrieval Infrastructure10 min
- Caching in GenAI Systems9 min
- Memory & Caching for Agent Systems8 min
- Memory, Databases & Caching: hands-on19 min
- Memory, Databases & Caching: scenarios15 min
07Module 7: Backend, FastAPI & Deployment7 parts · 96 min
Build, serve, deploy, and operate LLM-backed APIs with FastAPI.
- FastAPI Architecture9 min
- LLM Serving APIs10 min
- Deployment & Infrastructure12 min
- Production Operations11 min
- Calling LLM APIs Reliably23 min
- Backend, FastAPI & Deployment: hands-on18 min
- Backend, FastAPI & Deployment: scenarios13 min
08Module 8: GenAI System Design8 parts · 110 min
Design GenAI systems that scale, stay within budget, stay observable, and stay safe.
- Architecture Design for GenAI Systems14 min
- Scalability & Performance Engineering17 min
- Cost Optimization for GenAI Systems13 min
- Observability, Logging & Monitoring15 min
- Reliability, Safety & Guardrails13 min
- Data & Model Lifecycle Management6 min
- GenAI System Design: hands-on18 min
- GenAI System Design: scenarios14 min
09Module 9: AI Agents6 parts · 224 min
Understand agent architectures, frameworks, production concerns, and end-to-end use-case design.
- Agent Fundamentals & Architecture62 min
- Agent Frameworks & Ecosystem48 min
- Agent Production Systems36 min
- Agent Use-Case Design47 min
- AI Agents: hands-on18 min
- AI Agents: scenarios13 min
10Module 10: RAG9 parts · 223 min
Build RAG systems component by component, choose frameworks, run them in production, and apply them to real domains.
- RAG Components66 min
- Memory Management & Caching in RAG5 min
- RAG Frameworks & Tools36 min
- Production RAG37 min
- Legal & Compliance RAG17 min
- Healthcare Knowledge RAG17 min
- Enterprise Internal Knowledge Base RAG16 min
- RAG: hands-on17 min
- RAG: scenarios12 min
11Module 11: LLM Evaluation & Testing7 parts · 86 min
Build eval sets, use LLM-as-judge carefully, evaluate RAG and agents, and catch regressions in CI and production.
- Building Eval Sets14 min
- LLM-as-Judge in Practice15 min
- Evaluating RAG & Agents13 min
- Testing in CI & Regression11 min
- Online Evaluation & Feedback9 min
- LLM Evaluation & Testing: hands-on14 min
- LLM Evaluation & Testing: scenarios10 min
12Module 12: Serving, Latency & Cost Basics5 parts · 66 min
Understand where latency and cost come from and make LLM features faster and cheaper, including when self-hosting makes sense.
- Where Latency Comes From13 min
- Making It Faster & Cheaper14 min
- API vs Self-Hosted10 min
- Serving, Latency & Cost Basics: hands-on16 min
- Serving, Latency & Cost Basics: scenarios13 min
13Module 13: Multimodal & Voice AI5 parts · 64 min
Build with vision and document understanding, voice agents, and image generation APIs.
- Vision & Document Understanding16 min
- Voice Agents16 min
- Image Generation in Apps9 min
- Multimodal & Voice AI: hands-on13 min
- Multimodal & Voice AI: scenarios10 min
14Module 14: Safety & Security for Builders7 parts · 67 min
Ship LLM features safely: prompt injection, output handling, tool permissions, PII, secrets, guardrails and abuse limits.
- Prompt Injection in Apps You Build11 min
- Output Handling & Rendering Risks7 min
- Tool Permissions & Confirmations7 min
- PII, Secrets & Data Handling10 min
- Guardrails, Moderation & Abuse Limits8 min
- Safety & Security for Builders: hands-on13 min
- Safety & Security for Builders: scenarios11 min
15Module 15: Practical Build & Project Rounds7 parts · 116 min
Prepare for take-home builds, live coding, debugging rounds, project deep-dives and behavioral rounds.
- Take-Home Builds13 min
- Live Coding Tasks29 min
- Debugging Rounds15 min
- Project Deep-Dive15 min
- Behavioral for AI Engineers10 min
- Practical Build & Project Rounds: hands-on22 min
- Practical Build & Project Rounds: scenarios12 min
Reviews
to review this kit once you have finished it.