How modern AI works, one moving part at a time
Short, animated explainers of the ideas inside today's AI systems, from a single neuron to transformers, vector search and agents that call your tools. 48 topics, each with the original papers.
The journey of one prompt
Everything below is a part of this loop. Watch a request travel through it, then open any stage to see how it works.
Before the model: retrieve
Search your documents with a vector index (HNSW) and add the best passages to the prompt.
HNSW and RAGDuring: call tools
The model asks for a tool; an MCP server runs it on your systems and returns the result as context.
Model Context ProtocolAttention layers
Pick and repeat
Answer so far: The
Append, then run again for the next token, reusing cached attention.
Pick a chapter, or read in order
Each topic is a few minutes: an interactive visual, a plain explanation, why it matters for a product, and the primary sources.
Neural network basics
Neurons, layers, loss and how a network learns from its mistakes.
Language as numbers
How text becomes tokens, tokens become vectors, and meaning becomes distance.
Inside the transformer
Attention, positions, residual streams: the architecture behind every modern model.
Generating text
From probabilities to words, and the tricks that make it fast.
Speed and cost
Experts, quantisation and serving tricks that make large models affordable.
Training and alignment
Pre-training, fine-tuning and teaching a model what people actually want.
Retrieval and search
Keyword search, vector search, HNSW graphs and grounding answers in your data.
Agents and protocols
Models that plan, call tools and talk to your systems through MCP.
Observability
Traces, evals and monitoring: how you know a system works, in the lab and in production.
Interpretability and safety
Looking inside the model to see what it is doing, and why that matters.
A roadmap and a map
The roadmap turns the chapters into tracks with modules and a progress tick per topic. The topography draws every chapter, topic and term as one connected map.
How the ideas connect
Each line points from a topic to one worth reading first. Hover any topic to light up everything it builds on and everything it unlocks.
Want these ideas working on your data?
We build the retrieval, agents, voice and model systems these explainers describe. Start with a free discovery call.