├── 2024-06-24 - Workshop AI Programmer Handout.pdf └── README.md /2024-06-24 - Workshop AI Programmer Handout.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/go-go-golems/go-go-workshop/HEAD/2024-06-24 - Workshop AI Programmer Handout.pdf -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # LLMs for the working programmer 2 | 3 | Workshop time: June 25th, 9:00 AM. 4 | 5 | Youtube Video: https://www.youtube.com/watch?v=zwItokY087U 6 | 7 | Slack channel: #workshop-llm-working-programmer 8 | 9 | Participants need access to a LLM of their choice (recommended: gpt-4o and claude 3.5 sonnet). Please prepare a list of programming topics (projects you want to build, legacy code you might want to work on, frameworks and technologies you want to learn). I will provide an extensive list of ideas to try out during the talk. 10 | 11 | The handout to the talk will be added shortly before the workshop, I don't want to spoil the surprise. 12 | 13 | ## Prerequisites 14 | 15 | These are all optional and just FIY. 16 | 17 | ### LLMs 18 | 19 | I don't really know too much about the internals of models and how they are trained. So you don't need to either! 20 | 21 | - broad knowledge of how LLMs work (not need to go rabbit hole into the details) 22 | - transformer attention (autoregression in particular, the model uses its own output as part of generating further output) 23 | - tokens and context window size 24 | - RLHF instruct vs "pure" completion models / function calls and how they get encoded as special tokens 25 | 26 | ### Programming 27 | 28 | - [DSL design pattern](https://en.wikipedia.org/wiki/Domain-specific_language) . If you know common lisp or ruby you are well set. 29 | - a book that was fundamental to my education as a programmer and I think informs why I get so much out of LLMs: https://en.wikipedia.org/wiki/Paradigms_of_AI_Programming 30 | - [Blackboard System](https://en.wikipedia.org/wiki/Blackboard_system) a very effective pattern for at least conceptualizing how to build "agentic" systems (although I don't use agents and zero-shot for most of my use cases) 31 | - a really insightful programming with LLMs resources: https://simonwillison.net/ 32 | 33 | More general purpose books 34 | - [Computational Logic and Human Thinking - Artifically intelligent](https://www.amazon.com/Computational-Logic-Human-Thinking-Artificially/dp/0521123364) - in a way the most practical prompt engineering book I found 35 | - [Co Intelligence](https://www.amazon.com/Co-Intelligence-Living-Working-Ethan-Mollick/dp/059371671X) - only useful LLM book out there? 36 | --------------------------------------------------------------------------------