Embedded block-level LLMs (BYOM) with integrated Agentic RAG.
Play the motion
Pause
the animation
Play
(the animation is paused)
The route an AI request takes from a lesson block
A lesson is made of blocks, and an LLM is embedded in each block. A request leaves one block and reaches the model layer: Claude models, through Amazon Bedrock and Anthropic's API, or your own model (BYOM). Before it answers, the agentic RAG loop retrieves from a library. Each field has its own library: MedTech, FinTech, EdTech and LegalTech, so the answer is grounded in each field's own library. Every AI act then passes one authority gate, where people keep the authority, and lands in a record. The answer returns to the block. Arrowheads on the lines show the way the data moves: from the block to the model, round the loop and down to the library and back, on to the gate and the record, and back to the block.
Lesson blocks
An LLM in each block
Many models
Claude models or your own
Agentic RAG
Retrieves before it answers
Authority gate
People keep the authority
Record
Every AI act lands here
LLM
LLM
LLM
LLM
Claude models
through Amazon Bedrock
Claude models
through Anthropic's API
Your own model (BYOM)
MedTech
FinTech
EdTech
LegalTech
grounded in each field's own library
Answer returns to the block
The route an AI request takes from a lesson block
A lesson is made of blocks, and an LLM is embedded in each block. A request leaves one block and reaches the model layer: Claude models, through Amazon Bedrock and Anthropic's API, or your own model (BYOM). Before it answers, the agentic RAG loop retrieves from a library. Each field has its own library: MedTech, FinTech, EdTech and LegalTech, so the answer is grounded in each field's own library. Every AI act then passes one authority gate, where people keep the authority, and lands in a record. The answer returns to the block. Arrowheads on the lines show the way the data moves: from the block to the model, round the loop and down to the library and back, on to the gate and the record, and back to the block.
Lesson blocks
An LLM in each block
Many models
Claude models or your own
Agentic RAG
Retrieves before it answers
Authority gate
People keep the authority
Record
Every AI act lands here
LLM
LLM
LLM
LLM
Claude models
through Amazon Bedrock
Claude models
through Anthropic's API
Your own model (BYOM)
MedTech
FinTech
EdTech
LegalTech
grounded in each field's own library
Answer returns to the block