1. Decide what kind of AI person you want to be (for now)

• LLM App Builder – chatbots, copilots, agents, automations.
• GenAI Creative – images, videos, audio, content workflows.
• Infra / Systems Nerd – deployment, serving, evals, scaling.

Free resources to explore roles:

• DeepLearning.AI – “Generative AI for Everyone” (Andrew Ng) – a high-level overview, but great for understanding where different roles fit. https://www.deeplearning.ai/courses/generative-ai-for-everyone
• EvidentlyAI’s “Free LLM/GenAI courses in 2025” roundup – curated list of free LLM courses; skim to see which topics excite you (infra vs apps vs creativity). https://www.evidentlyai.com/blog/llm-genai-courses

2. Go LLM-first, not “math-first”

You spend 1–2 weeks learning how LLMs behave:

• Prompts: system vs user vs tools.
• Temperature / max tokens / JSON mode.
• Patterns: role prompting, few-shot, chain-of-thought, rewriting, and extraction.

Free resources:

• Google – “Introduction to Large Language Models” – short, free intro to how LLMs work + prompt basics. https://machinelearningmastery.com/5-free-courses-mastering-llms/
• Hugging Face LLM Course – Chapter 1 (What are LLMs & Transformers) – free, hands-on explanation with code, but still beginner-friendly. https://huggingface.co/learn/llm-course/en/chapter1/1
• Andrej Karpathy – “Intro to LLMs” (YouTube) – very current mental model of how these models really work. https://www.reddit.com/r/LLMDevs/comments/1ieks86/free_resources_for_learning_llms/

Practice prompts like:

• “Turn this messy meeting note into 5 clear action items + owners.”
• “Extract a JSON list of {company, role, location} from this job post.”