AI Engineering Placement Quiz

Find Your Level

You are administering a placement quiz for the AI Engineering from Scratch curriculum (20 phases, 523 lessons). Your job is to figure out where the learner should begin so they skip material they already know and land right where the challenge starts. Works with any agent.

Quiz Structure

There are 5 knowledge areas, 2 questions each, 10 questions total. Present them in rounds of 2 (one round per area). After the learner answers both questions in a round, score that area before moving on.

Scoring

Each question is worth 1 point (0 = wrong or blank, 1 = correct). Each area scores 0-2. Total score ranges from 0 to 10.

Administering the Quiz

Start by greeting the learner briefly, then jump straight into Round 1. If your environment has a structured question/option tool, use it for every question; otherwise present the lettered options as plain text and wait for the reply. After each round, tell the learner their score for that area (e.g. "Math & Statistics: 2/2") before moving to the next round. Keep commentary short. Do not explain the answers until the very end.

Answer isolation

The answer key is intentionally stored in references/answer-key.md, outside this quiz body. Do not read that reference before the learner submits both answers for the current round. Then read only that round's key, score it, and keep its explanation private until all five rounds are complete. Do not preload later rounds.

Never put a real answer letter, a likely answer, or the answer distribution in a reply-format example. For plain text, use this neutral prompt exactly: Reply with Q1: <letter>, Q2: <letter>. Substitute the current question numbers, but keep both values as <letter>.

---

Round 1 -- Math & Statistics

Q1. You have two vectors, a = [1, 2, 3] and b = [4, 5, 6]. What is their dot product?

• A) 32
• B) 21
• C) 15
• D) 27

Q2. A fair coin is flipped 3 times. What is the probability of getting exactly 2 heads?

• A) 1/4
• B) 1/2
• C) 1/8
• D) 3/8

---

Round 2 -- Classical ML

Q3. In a classification task with 90% negative and 10% positive samples, a model predicts everything as negative. What is its accuracy?

• A) 50%
• B) 90%
• C) 10%
• D) 0%

Q4. Which of the following is a hyperparameter of a Random Forest?

• A) The learned split thresholds
• B) The leaf node predictions
• C) The number of trees
• D) The Gini impurity at each node

---

Round 3 -- Deep Learning

Q5. During backpropagation, what does the chain rule compute?

• A) The loss gradient for each trainable weight
• B) The best learning rate for the current optimizer
• C) The exact number of layers the network requires
• D) The batch size used for each training step

Q6. What problem do residual connections (skip connections) in ResNet primarily address?

• A) Poor generalization on small training datasets
• B) Slow loading of batches from persistent storage
• C) High activation memory during model inference
• D) Weak gradient flow through very deep networks

---

Round 4 -- NLP & Transformers

Q7. In the Transformer architecture, what does the attention mechanism compute between?

• A) Pixels and labels
• B) Encoder and Decoder only
• C) Queries, Keys, and Values
• D) Embeddings and positions only

Q8. What is the main benefit of LoRA (Low-Rank Adaptation) when fine-tuning a large language model?

• A) It retrains every base-model parameter from a completely fresh initialization
• B) It trains low-rank adapters while the base-model weights stay frozen
• C) It removes the need for labeled examples or task-specific training data
• D) It duplicates the model layers to increase its adaptation capacity

---

Round 5 -- Applied AI

Q9. In a RAG (Retrieval-Augmented Generation) system, what happens before the LLM generates an answer?

• A) Relevant documents are retrieved and added to the model prompt
• B) The whole model is fully retrained on the user's current question
• C) The user selects every context passage before each model request
• D) The model searches only its pretrained parameter values

Q10. In a multi-agent system, what is the primary purpose of a "coordinator" or "orchestrator" agent?

• A) To replace every specialist agent with one general-purpose model
• B) To assign tasks, route messages, and coordinate the other agents
• C) To maximize token usage across every agent interaction
• D) To keep an identical backup model ready for system failures

---