How This Course Continues in Phase 2
You'll learn to
- -Know exactly what changes in Phase 2: real hands-on practice via the GenAI Lab
- -Check yourself against a one-sentence recap of each Phase 1 idea
- -Know which Phase 1 ideas Phase 2 leans on most, and where to review them
- -Feel ready to continue directly into Phase 2's first chapter
Congratulations. Finishing this module means you have finished all of Phase 1. Every concept from here forward in this course builds directly on the foundation you just built: vectors and distance, probability and loss, gradient descent and backpropagation, overfitting and evaluation, and the discriminative and generative distinction from the last two chapters.
A One-Sentence Check on Phase 1
Before you move on, see whether you can say each of these in your own words. If one is hazy, the chapter that introduced it is a short revisit:
- -Vectors and distance: a vector is a list of numbers that describes something, and distance tells you how alike two things are.
- -Probability and loss: probability says how likely something is, and loss is one number for how wrong the model is, which training tries to shrink.
- -Gradient descent and backpropagation: nudge every parameter in the direction that lowers the loss, and backpropagation is the efficient way to work out those nudges layer by layer.
- -Overfitting and evaluation: a model that memorizes its training data fails on new data, so you always measure on data it has not seen.
- -Discriminative versus generative: a discriminative model learns the boundary between classes, while a generative model learns how the data itself is produced, so it can create new examples.
What Changes Starting Now
The biggest practical change is that from this point on, nearly every module ends with a direct bridge into a real, hands-on level in the GenAI Lab. Phase 1 had no lab exercises of its own, because the lab is entirely about generative AI systems, but everything Phase 2 covers has a matching level waiting for you to actually build it, not just read about it.
- -Phase 2 starts with LLM foundations: tokenization, embeddings, and memory, the concrete mechanics behind the "how a transformer processes language" preview from the Neural Networks module.
- -From there: retrieval-augmented generation, agents and tool use, and finally production and LLMOps concerns.
- -The course closes with a capstone module of fully worked GenAI system design examples, mirroring the worked examples you may have seen in ScaleDojo's other Learn courses.
If any Phase 1 concept feels shaky, especially vectors and embeddings, loss functions, or the discriminative versus generative distinction from the last two chapters, this is a good moment to skim back over it. Phase 2 assumes this vocabulary rather than re-explaining it from scratch.
How to Approach Phase 2
Phase 2 rewards building as you read. Read a chapter, then go and build the level it points to, and let the lab show you which parts you really understood. If a lab level exposes a gap, come back and reread the chapter: the second reading will be quicker and will stick.
Ready when you are. Continue on to Welcome to Generative AI Engineering, the first chapter of Phase 2.
Interview Signal is part of Pro
See a real weak answer next to a real strong one for this exact topic.
Quiz is part of Pro
Test what you just read with a short quiz, and bank the XP.
Discussion0
Join the Discussion
Sign in to leave comments, reply to others, or like insights.
No comments yet. Be the first to start the thread!