Skip to content

ankityddv/AIFromScratch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AIFromScratch

my personal notebook for learning ai from scratch.

this repo is not meant to be a polished course or tutorial.

it's a collection of my notes, learnings, resources, plans, experiments, observations, and project ideas as i progress through the ai/ml space.

the goal is simple:

  • understand concepts deeply
  • build strong fundamentals
  • keep track of what i'm learning
  • create a reference i can revisit in the future
  • document projects and experiments along the way

what you'll find here

terminology & concepts

notes explaining fundamental concepts in simple language.

examples:

  • tokens
  • parameters
  • context windows
  • transformers
  • embeddings
  • attention
  • fine-tuning
  • quantization
  • rag
  • agents

learning roadmap

a living document containing:

  • what i'm currently learning
  • what i've completed
  • what i'm planning to learn next

this will evolve as my understanding grows.


resources

a collection of useful learning material including:

  • youtube playlists
  • courses
  • blogs
  • research papers
  • github repositories
  • books
  • documentation

for each resource i may also add:

  • notes
  • takeaways
  • timestamps
  • things worth revisiting later

study notes

notes taken while watching videos, reading blogs, exploring repositories, or following courses.

these notes are written for future me.

the goal is not perfect documentation.

the goal is quick understanding when revisiting a topic months later.


experiments

small hands-on experiments to better understand concepts.

examples:

  • tokenization experiments
  • embedding experiments
  • transformer implementations
  • training toy models
  • inference experiments
  • rag prototypes
  • agent workflows

project ideas

a collection of ideas that i may explore in the future.

each idea may include:

  • problem statement
  • rough architecture
  • learning objectives
  • implementation notes

progress tracker

a simple log of:

  • what i'm learning
  • completed topics
  • active topics
  • upcoming topics

this helps me stay consistent and measure progress over time.


structure

.
├── README.md
├── roadmap
│   └── learning-plan.md
├── notes
│   ├── terminology.md
│   ├── transformers.md
│   ├── embeddings.md
│   ├── rag.md
│   └── agents.md
├── resources
│   └── resources.md
├── experiments
├── projects
└── progress
    └── learning-log.md

current focus

  • building strong fundamentals
  • understanding how llms work internally
  • learning transformers from first principles
  • understanding training vs inference
  • learning how modern ai applications are built

philosophy

learn slowly.

understand concepts instead of memorizing them.

focus on fundamentals first.

build things whenever possible.

document everything worth revisiting.

future me should be able to open this repo and quickly understand what i learned, why i learned it, and where i left off.

About

my personal ai learning notebook. a collection of notes, terminology, learning plans, resources, experiments, project ideas, and progress tracking as i learn ai, ml, llms, slms, transformers, rag, agents and related topics from scratch.

Resources

Stars

Watchers

Forks

Contributors