Intermediate 6 topic areas 71+ exercises

AI / ML Engineer

AI/ML engineers build systems that learn from data. This path covers the vocabulary for discussing model architectures, training pipelines, evaluation metrics, and the rapidly evolving language of LLMs, embeddings, and RAG — used daily in design reviews, paper discussions, and stakeholder updates.

Start first exercise → Browse all exercises

Topics covered

ML fundamentals
LLMs & foundation models
RAG & embeddings
MLOps & deployment
Model evaluation
Responsible AI

Vocabulary spotlight

4 terms every AI / ML Engineer should know in English:

hallucination n.

When an LLM generates plausible-sounding but factually incorrect information

"The model hallucinated a citation — always verify AI-generated references."

fine-tuning n./v.

Adapting a pre-trained model to a specific task by training it on domain-specific data

"We fine-tuned GPT on our support tickets to improve its product knowledge."

embeddings n.

Vector representations of text that capture semantic meaning for similarity search

"We store document embeddings in a vector database for semantic retrieval."

context window n.

The maximum amount of text an LLM can process in a single inference call

"The 200k-token context window lets us pass the entire codebase to the model."

Open full glossary →

📚 Vocabulary Reference

Key terms organised by category for AI / ML Engineers:

ML Fundamentals

training setvalidation settest setoverfittingunderfittinggradient descentloss functionepochbatch sizelearning rate

LLMs & Transformers

transformerattention mechanismtokenisationcontext windowprompttemperaturetop-psystem promptinstruction tuningRLHF

RAG & Retrieval

RAGvector databaseembeddingssemantic searchchunkingretrievalgroundinghallucinationfaithfulnessrelevance

MLOps

model registryfeature storedata pipelineexperiment trackingmodel versioningshadow deploymentA/B testdrift detectionretraining trigger

Study full vocabulary modules →

Recommended exercises

AI & ML Vocabulary 30 exercises

Vocabulary

AI Prompting Exercises 15 exercises

Speaking/Writing

Writing Model Cards & System Design Docs 5 exercises

Writing

Writing Architecture Records & API Docs 6 exercises

Writing

Tech-to-Business: Explaining Model Behaviour 10 exercises

Speaking

AI/ML Engineer Interview Questions 5 exercises

Interview

Real-world scenarios you'll practise

Explaining model evaluation metrics to a non-technical product manager
Presenting RAG architecture trade-offs in a system design review
Writing an ML model card for internal governance
Discussing responsible AI concerns with a compliance officer

🎯 Interview questions specific to this role

Practise answering these questions out loud — or in writing. Each question targets a real interviewer concern for AI / ML Engineers.

What is the difference between a transformer and an RNN?
How do you evaluate the quality of an LLM-based application?
What is retrieval-augmented generation and when would you use it?
How do you detect and mitigate bias in a machine learning model?
Walk me through your MLOps process from model training to production deployment.

Practice all interview exercises →