AI Whiteboard #1: The Architecture of Modern AI

​Forget the hype. This session is about understanding how modern AI actually works.

​Join us for a 2 hour whiteboard-only meetup where we'll break down the complete architecture behind today's AI systems from the fundamentals of neural networks to transformers, embeddings, attention mechanisms, and Large Language Models (LLMs). No slides, no coding, just diagrams, discussions, and first-principles thinking.

​Whether you're a student, software engineer, founder, researcher, or simply AI-curious, you'll leave with a clear mental model of the technology powering tools like ChatGPT, Claude, Gemini, and other modern AI systems.

​What you'll learn

• ​Why AI has become so powerful in recent years

• ​The evolution from Machine Learning → Deep Learning → Transformers

• ​Neural Networks explained from first principles

• ​Tokens: how AI reads and processes text

• ​Embeddings: how language becomes numbers

• ​The Attention mechanism—why it changed everything

• ​Transformer architecture explained visually

• ​Training vs Inference

• ​What makes an LLM a "Large Language Model"

• ​Context windows, parameters, and reasoning

• ​Hallucinations: why they happen

• ​How tools like ChatGPT, Claude, and Gemini actually work

• ​Where RAG, AI Agents, and MCP fit into the modern AI stack (high-level overview)

​Who should attend?

• ​Students exploring AI

• ​Software engineers

• ​Startup founders

• ​Product managers

• ​Anyone who wants to understand AI beyond prompts and buzzwords

​No prior AI experience is required. Just bring your curiosity—we'll build everything from the ground up on the whiteboard.