Large Language Models (LLMs) and Retrieval Augmented Generation (RAG)
Master Large Language Models (LLMs) and build sophisticated text generation applications. Learn prompt engineering, model selection and cost optimization, and dive deep into Retrieval-Augmented Generation (RAG) with vector databases, then evaluate performance with RAGAS and build an end-to-end RAG application.
What you'll learn
- Understand LLM architectures, tokenization, attention, and core capabilities
- Build stateful LLM-powered chatbots with effective prompt engineering
- Control and refine LLM behavior using inference parameters and instruction refinement
- Implement tokens, embeddings, and vector search for semantic search systems
- Design and build complete RAG workflows using vector databases like ChromaDB
- Evaluate RAG systems with metrics such as context precision, recall, faithfulness, and answer relevancy using RAGAS
- Apply strategic model selection and cost economics, including hybrid routing and TCO analysis
- Complete a capstone project building a NASA mission RAG-based chat system
Skills you'll gain
- Explain core concepts of large language models and their capabilities
- Build and deploy stateful LLM chatbots using system, user, and assistant messages
- Design and refine prompts using role, task, context, examples, and output format
- Tune inference parameters such as temperature, top-p, max tokens, and stop sequences
- Implement tokenization, embeddings, and vector search for semantic retrieval
- Select and combine LLMs based on performance, speed, cost, and control (TCO)
- Design and implement a full Retrieval-Augmented Generation (RAG) workflow
- Use vector databases (e.g., ChromaDB) with metadata filters for RAG pipelines
- Engineer prompts that ground answers in retrieved context with citations
- Evaluate and improve RAG systems using RAGAS and metrics like context precision, recall, faithfulness, and answer relevancy
- Build an end-to-end RAG chatbot for NASA mission intelligence data
Prerequisites
- • Deep learning
- • Generative AI Fluency
- • PyTorch
- • Hugging Face
- • Intermediate Python
- • Ability to communicate fluently and professionally in written and spoken English
Who this course is for
- → Developers and engineers who want to build LLM-powered applications
- → Data scientists and ML practitioners looking to implement RAG systems
- → Technical professionals seeking to optimize LLM cost, performance, and reliability
Provider
Udacity
Related Courses
Python Bootcamp
Python Bootcamp covers fundamentals of Python programming, including control structures, advanced data types, functions, modules, packages, multithreading, exception handling, file handling, GUI design, and database connectivity, preparing learners for future work in data science and machine learning.
Systems Engineering
This course introduces Systems Engineering principles across the lifecycle of complex systems, covering system design, architecture, requirements analysis, modeling, verification, lifecycle models (Waterfall, V-Model, Spiral, Agile), SysML, risk management, trade-off analysis, and a Smart Home Security System project.
IT Systems Design and Analysis
Prepare to design, analyze, and evaluate IT systems using data flow diagrams, ERDs, UML, and feasibility analysis. Learn to assess existing systems, identify inefficiencies, compare solution alternatives, and deliver a digital transformation strategy through a hands-on final project.
Hands On FullStack Development Course with Infrastructure Management Product implementation
A 180-day, project-first full-stack infrastructure course where you build and operate production-grade services with CI/CD, testing, observability, and operational playbooks, aimed at taking you from toy projects to real-world deployment experience.
Learn Typescript
Hands-on introduction to TypeScript fundamentals and their application in real projects. Learn core typing concepts, advanced TypeScript features, and how to use TypeScript with React and Express while building safer, more maintainable JavaScript applications.
AI Engineering Course
Designed to help software engineers transition to AI engineering, with detailed breakdowns of vector databases, indexing, large language models, attention, and core optimizations so you can understand how LLMs work and use them to build real-world applications.