Mini Map
Search
⌘ K
AI Features
Log In
The AI Engineer Interview Crash Course*
1.
How Models Work
The Neural Network Engine
ReLU vs Sigmoid vs Softmax
Forward Pass Explained
How Backpropagation Works
Highlight Card
Gradient Descent Variants
Why AdamW is the Standard
Test Your ML Foundations
Transfer Learning vs From Scratch
Why Text Must Become Numbers
Character vs Word vs Subword
How BPE Tokenization Works
Embeddings as Geometry of Meaning
Static vs Contextual Embeddings
Sparse vs Dense Retrieval
Tokenization Knowledge Check
Why Self-Attention Replaced RNNs
Q, K, V — The Attention Trio
Highlight Card
Multi-Head Attention Explained
Self-Attention vs Cross-Attention
GQA — The Modern LLM Default
Attention Mechanism Check
Why Transformers Need Position Info
Causal Mask vs Padding Mask
Layer Norm Over Batch Norm
Flash Attention's Key Insight
True / False
How Models Work — Recap
2.
Training, Optimization, and Scale
Why Inference Matters Most
Autoregressive Generation Explained
Greedy vs Beam vs Sampling
Three Sampling Knobs
The KV-Cache Explained
Speculative Decoding Speedup
Highlight Card
Check Your Decoding Knowledge
From Base Model to Assistant
Pretraining vs Fine-Tuning
Why SFT Alone Falls Short
How RLHF Works
RLHF vs DPO
Constitutional AI and RLAIF
Highlight Card
Alignment Pipeline Check
The VRAM Problem
Full Fine-Tuning vs PEFT
How LoRA Works
Highlight Card
Quantization Cuts Model Size
Knowledge Distillation
Compression Strategy Quiz
Scaling Laws Set Training Budgets
Mixture of Experts Architecture
Highlight Card
Autoregressive vs Diffusion Models
Enabling Million-Token Contexts
Architecture Trade-Off Quiz
Evaluation Has No Perfect Metric
BLEU vs ROUGE
Highlight Card
Chatbot Arena and Elo Ratings
LLM-as-Judge and Its Biases
True / False
Fill in the Blank
Training, Optimization, Scale
3.
Applied AI Engineering
Prompting is Programming
What is In-Context Learning?
Highlight Card
Chain-of-Thought Prompting
True / False
How System Prompts Work
Prompt Injection Explained
Scenario Card
How Function Calling Works
Prompting Knowledge Check
Why LLMs Hallucinate
The RAG Pipeline
RAG vs Fine-Tuning
Chunking Strategy Matters
Dense vs Sparse Retrieval
Re-Ranking for Precision
Highlight Card
Scenario Card
RAG Pipeline Check
Agent vs Chatbot
ReAct — Think Then Act
MCP — USB-C for AI Tools
MCP vs A2A
Multi-Agent Failure Modes
Fill in the Blank
Top LLM Safety Risks
Bias in LLMs
Highlight Card
Mechanistic Interpretability
Production LLM Best Practices
Scenario Card
Safety and Production Check
Applied AI Engineering Recap
Claim your Certificate
Home
Courses
The AI Engineer Interview Crash Course*
Speculative Decoding Speedup
We'll cover the following...
This component is not supported in the current context
...