For advanced participants with basic knowledge of machine learning, LLMs, and Python.

Cohort 4 - Intermediate to Advanced

Use advanced prompting techniques and tools to improve the capabilities, performance, and reliability of LLM-powered applications

Elvis Saravia

Advanced Prompt Engineering for LLMs

Google

Amazon

Microsoft

LinkedIn

Airbnb

OVERVIEW OF THE COURSELLMs (Large Language Models) show powerful capabilities, but not knowing how to effectively and efficiently use them often leads to reliability and performance issues.&nbsp;Prompt engineering helps to improve discover capabilities, improve reliability,&nbsp;reduce failure cases, and save on computing costs when building with LLMs. This is a hands-on, technical course that teaches how to effectively build with LLMs. It covers the latest prompting techniques (e.g., fe-shot, chain-of-thought, RAG, prompt chaining) that you can apply to a variety of complex use cases such as building personalized chatbots, LLM-powered agents, prompt injection detectors, LLM-powered evaluators, and much more. Topics include: • Taxonomy of Prompting Techniques• Tactics to Improve Reliability• Structuring Inputs and LLM Outputs• Zero-shot/Few-shot/Many-Shot Prompting• Chain of Thought Prompting• Self-Reflection &amp; Self-Consistency• Meta Prompting &amp; Automatic Prompt Engineering• ReAcT Prompting Framework• Retrieval Augmented Generation (RAG)• Fine-Tuning LLMs• Function Calling &amp; Tool Usage• LLM-Powered Agents (Agentic Workflows)• LLM Evaluation &amp; Judge LLMs• AI Safety &amp; Moderation Tools• Adversarial Prompting (Jailbreaking and Prompt Injections)• Common Real-World Use Cases of LLMs• Overview of LLM Tools... and much more PREREQUISITES• This is a technical course and you need to be knowledgeable in Python to take this course.• Basic knowledge of LLMs is beneficial but not required. If you don't have experience using Python, we recommend our beginner's course: https://maven.com/dair-ai/llms-for-everyone ABOUT THE INSTRUCTORElvis, the instructor for this course, has vast experience doing research and building with LLMs and Generative AI. He is a co-creator of the Galactica LLM and author of the popular Prompt Engineering Guide. He has worked with world-class AI teams like Papers with Code, PyTorch, FAIR, Meta AI, Elastic, and many other AI startups. Reach out to training@dair.ai for any questions, corporate trainings, and group/student discounts. WHO THE COURSE HAS HELPEDThis course has helped AI startups freelancers, and professionals at companies like Apple, Microsoft, Google, LinkedIn, Amazon, Coinbase, Asana, Airbnb, Intuit, JPMorgan Chase &amp; Co, and many others.

Effectively prompting and building with LLMs

Developers building applications on top of LLMs

Builders wanting to improve LLM reliability, efficiency, and performance for their LLM-powered applications.

Professionals interested in leveling up on how to better use and apply LLMs.

Who is this course for

What you’ll get out of this course

<ul><li>Learn key elements and tactics for designing effective prompts</li><li>Design, test, and optimize prompts to improve model performance and reliability for different tasks such as text summarization and information extraction</li></ul>

Design and optimize prompts

<ul><li>Review and apply the latest and most advanced prompt engineering techniques (few-shot learning, chain-of-thought, RAG, prompt chaining, self-consistency, self-verification, etc.)</li><li>Build with approaches like ReAct, RAG, function calling, and LLM-powered agents</li></ul>

Build a robust framework to effectively apply advanced prompt engineering techniques

<ul><li>Develop use cases such as tagging systems, personalized chatbots, evaluation systems, product review analyzers, and more</li><li>Build advanced applications that involve combining knowledge with conversational assistants and using LLMs with external tools and knowledge</li></ul>

Develop use cases and build applications

<ul><li>Design a robust framework for evaluating and measuring the quality, diversity, safety, and robustness of LLMs</li><li>Compare prompt engineering, RAG, and fine-tuning</li><li>Cover safety topics like prompt injection and moderation tools</li></ul>

Perform evaluations for your applications

<ul><li>Review the latest prompt engineering and LLM tools such as ChatGPT, Llama Index, Comet, LangChain, Flowise, Scale AI's Spellbook, and many more</li><li>Discuss current trends, papers, and future directions in prompt engineering</li></ul>

Learn prompt engineering tools

<ul><li>Discuss the key elements of designing an effective prompt&nbsp;</li><li>Design prompts for common tasks such as question answering and information extraction</li><li>Apply a taxonomy of popular prompting techniques such as role-playing, zero-shot, and few-shot learning </li></ul>

Structuring Effective Prompts

<ul><li>Apply advanced prompting techniques like prompt chains, chain-of-thought, ReAct, and RAG to improve the performance of LLMs&nbsp;</li><li>Improve model reliability, efficiency, and performance</li><li>Learn cost-effective and efficient prompting tactics</li><li>Learn from a case study about a data-augmented question-answering tool</li></ul>

Advanced Prompting Techniques & LLM Reliability

<ul><li>Discuss prompting techniques and approaches for assessing model safety, toxicity analysis, mitigating bias, reducing hallucination, and testing prompt injections</li><li>Applying different approaches for evaluating LLMs for tasks such as text classification and summarization</li><li>Evaluate and compare prompt engineering techniques with fine-tuned models</li></ul>

Evaluating LLMs & AI Safety

<ul><li>Review the latest tools and best practices for prompt engineering to effectively build with language models</li><li>Discuss end-to-end, real-world use cases and applications, such as combining knowledge with conversational bots and using LLM with external tools</li><li>Discuss current papers, trends, recommendations, and future directions</li></ul>

Prompting Tools, Advanced Use Cases, & Applications

Course syllabus

What people are saying

Meet your instructor

Director, Data Science at Fidelity Investments

4 live sessions (includes lectures, demos, exercises, and projects)

Live Sessions

Optional office hours to ask questions and receive guidance related to the course topics

Live Office Hours

Book a free 1-on-1 session with the instructor to further discuss careers, products, use cases, or anything related to building with LLMs.

1-on-1 Sessions

Includes additional readings and self-paced tutorials + bonus exercises to practice prompt engineering techniques and tools for different use cases and applications

Bonus content

Course schedule

This course builds on live workshops and hands-on projects

Active hands-on learning

You’ll be interacting with other learners through breakout rooms and project teams

Interactive and project-based

Join a community of like-minded people who want to learn and grow alongside you

Course completed by

About this course

Get notified about the next cohort

Learn from live sessions