Hi everyone, during a talk with Marco I got the idea that it might be helpful to lay down the areas of expertise needed to become an AI pro. (just focusing on LLM, rather than DS, ML, ...) Or rather the learning paths, because there are different roles to take. This is how I see it. Dave - What do you think? : @Dave Ebbelaar @Mamatha Naganna @Marco Bottaro @Ana Crosatto @Brandon Phillips A] AI Tech Knowledge ------------------------------- (in the order of going down the rabbit hole) 1) LLM Ecosystem / platform user (SaaS) Become proficient with ChatGPT, Bard, Claude Understand prompting, plugins, GPTs, ...; knowing some "super prompts" Also experimental Agent platforms like SuperAGI, AutoGPT, ... 2) LLM SaaS application user (SaaS) Learn about all the different end-user tools out there and use them in your area (video, presentations, translations, ...) 3) LLM NoCode application coder (PaaS) Web/Desktop: Learn about flowise, open gpt, open ai assistent (?) Cloud-Designers: Bedrock (?), Azure OpenAI Studio, Azure AI Studio 4) LLM Code application coder (the a16z essay is a good way to understand it) LLM frameworks and python libraries (langchain, autogen, ...), python knowledge Prompting, Vector Databases Coding environments (Desktop: VS Code) or alternatively Cloud-based Optional: everything else and non-AI related to building an application: web development, app development, UX-Design, design, ... 5) LLM model developers Fine-tuning models, Olama, ... B] Subject matter knowledge - Industry and Job role knowledge - applying the knowledge in your industry or job -------------------------------------------------------------------------------------------------------------------------------- Application of AI in your industry Application of AI for your job role / profession (agile coach, developer, business analyst, ...)