Spécialisation Generative AI Engineering with LLMs

Explain how one-hot encoding, bag-of-words, embeddings, and embedding bags transform text into numerical features for NLP models

Implement Word2Vec models using CBOW and Skip-gram architectures to generate contextual word embeddings

Develop and train neural network-based language models using statistical N-Grams and feedforward architectures

Build sequence-to-sequence models with encoder–decoder RNNs for tasks such as machine translation and sequence transformation

Explain the role of attention mechanisms in transformer models for capturing contextual relationships in text

Describe the differences in language modeling approaches between decoder-based models like GPT and encoder-based models like BERT

Implement key components of transformer models, including positional encoding, attention mechanisms, and masking, using PyTorch

Apply transformer-based models for real-world NLP tasks, such as text classification and language translation, using PyTorch and Hugging Face tools

Sought-after, job-ready skills businesses need for working with transformer-based LLMs in generative AI engineering

How to perform parameter-efficient fine-tuning (PEFT) using methods like LoRA and QLoRA to optimize model training

How to use pretrained transformer models for language tasks and fine-tune them for specific downstream applications

How to load models, run inference, and train models using the Hugging Face and PyTorch frameworks

In-demand generative AI engineering skills in fine-tuning LLMs that employers are actively seeking

Instruction tuning and reward modeling using Hugging Face, plus understanding LLMs as policies and applying RLHF techniques

Direct preference optimization (DPO) with partition function and Hugging Face, including how to define optimal solutions to DPO problems

Using proximal policy optimization (PPO) with Hugging Face to build scoring functions and tokenize datasets for fine-tuning

In-demand, job-ready skills businesses seek for building AI agents using RAG and LangChain in just 8 hours

How tapply the fundamentals of in-context learning and advanced prompt engineering timprove prompt design

Key LangChain concepts, including tools, components, chat models, chains, and agents

How tbuild AI applications by integrating RAG, PyTorch, Hugging Face, LLMs, and LangChain technologies

Gain practical experience building your own real-world generative AI application to showcase in interviews

Create and configure a vector database to store document embeddings and develop a retriever to fetch relevant segments based on user queries

Set up a simple Gradio interface for user interaction and build a question-answering bot using LangChain and a large language model (LLM)