Responsible Generative AI and Local LLMs

1h 48mIntermediate2024-07-16

Authors

Noah Gift

MLOps Expert | Solopreneur | Author | Adjunct Professor | CTO

Course details

In this course, MLOps expert Noah Gift guides you through mastering the theory of Generative AI and using local LLMs like Mistral, llamafile, Candle, and Lorax. Learn about profit sharing concepts, the tragedy of the commons, and game theory as it relates to generative AI. Dive into perfect competition, negative externalities, and how to fine tune Mistral with Ludwig. Find out how to get started with llamafile and how to execute Hello World in Candle. Explore Rust and Candle, as well as how to transcribe audio with Whisper. Plus, learn how to use Rust for LLMs.

Skills covered

Responsible AINatural Language Processing (NLP)Generative AIArtificial Intelligence (AI)One-Off

Concepts

1. Responsible Generative AI and Local LLMs

01 - Coding ELO in Python
02 - Coding ELO in Rust
03 - Coding ELO in R
04 - Coding ELO in Julia
05 - Profit sharing concepts
06 - Tragedy of the commons
07 - Deploying LLMs with Lorax and SkyPilot
08 - Fine-tune Mistral and Ludwig
09 - Game theory in generative AI
10 - Perfect competition
11 - Negative externalities
12 - Regulatory entrepreneurship
13 - Creating reinforcement bias
14 - Getting started with Mozilla llamafile
15 - Developing cosmopolitan
16 - Building blocks for generative AI with whisper.cpp
17 - Transcribing with Whisper
18 - Portable phrase CLI
19 - Candle hello world
20 - Exploring StarCoder in Rust
21 - Whisper Candle transcriber
22 - Local system metrics
23 - Exploring remote development on AWS
24 - Rust for large language models (LLMs)
25 - The continuous build binary
26 - Serverless inference
27 - Rust CLI inference
28 - Rust chat inference
29 - The chat loop with StarCoder
30 - Invoke an LLM on an AWS G5 instance, part 1
31 - Invoke an LLM on an AWS G5 instance, part 2