AI Learning

Notes from learning how modern AI systems actually work — the mechanics under the abstractions.

Showing 8 of 8 AI notes

RAG2

RAG
intermediate
RAG vs Fine-tuning
Two different mechanisms for adapting an LLM to your domain, what each one actually changes inside the system, and how to pick between them.
- RAG
- Fine-tuning
- LLM
- +1
11 min read
RAG
beginner
What is RAG?
How retrieval augmented generation grounds LLM outputs in your own data, and where the pattern breaks down in production.
- LLM
- RAG
- Embeddings
- +1
10 min read

Embeddings2

Embeddings
intermediate
Vector Databases Explained
Why nearest-neighbor search needs its own index structure, how HNSW works, and what actually differs between vector database options.
- Vector Databases
- Embeddings
- Databases
- +1
12 min read
Embeddings
beginner
What are Embeddings?
How text becomes a vector, why cosine similarity works as a proxy for meaning, and what actually determines embedding quality.
- Embeddings
- LLM
- Vector Databases
- +1
9 min read

Agents2

Agents
intermediate
Tool Calling
How structured tool use actually works under the hood, why it isn't function execution, and what makes a tool definition easy for a model to call correctly.
- Tool Calling
- Agents
- API Design
- +1
10 min read
Agents
intermediate
AI Agents Explained
What separates an agent from a single LLM call, how the agent loop actually runs, and where autonomous systems fail in practice.
- Agents
- Tool Calling
- LLM
- +1
12 min read

Prompt Engineering1

Prompt Engineering
beginner
Prompt Engineering Fundamentals
Structuring prompts, choosing between system and user content, and the failure modes that show up once a prompt leaves the playground.
- Prompt Engineering
- LLM
- Evaluation
- +1
11 min read

Evaluation1

Evaluation
advanced
LLM Evaluation
Building eval sets, choosing between deterministic checks and model graders, and catching regressions before a prompt change ships.
- Evaluation
- LLM
- Observability
- +1
14 min read