less than 1 minute read

Day 1: Understanding the Stack

Starting my 120-day LLM Systems hands-on journey. The goal is to deeply understand every layer of building, fine-tuning, and deploying LLM-powered systems in production.

Today’s Focus

  • The LLM stack overview: tokenization, embeddings, attention, decoding
  • Key architectural choices (decoder-only vs encoder-decoder)
  • Setting up a local inference pipeline

Key Takeaways

Update this section with your learnings.

Resources

  • Add links to papers, tutorials, or repos you referenced.

This is part of my 120 Days of LLM Systems challenge.