liveBook · Manning

Execution graph

As explained in Investing for Programmers

An execution graph is a graph representation of workflow stages and the dependencies that determine their execution order.

An execution graph represents workflow stages and the dependencies that determine their order. In retrieval-augmented generation, it orchestrates retrieval first and response generation second, with shared state carrying the question, context, and answer. section 9.3.2

RAG execution

A question about autonomous-driving companies enters the graph. Retrieval supplies matching documents; generation uses that context to produce the answer. The compiled graph can be invoked with the question and its answer read from the resulting state.

How Investing for Programmers covers it

How deep it goes
Defines shared state and two stages, then shows sequencing, compilation, visualization, and invocation without discussing scheduling internals.
Examples
Uses autonomous-driving company questions, vector-store retrieval, prompt generation, graph visualization, and printed answers.
What it assumes
Assumes prepared documents, an in-memory vector store, embeddings, a prompt, and an initialized language model.
What you can do after
Build it from scratch with LangGraph by defining state, sequencing stages, compiling the graph, and invoking it with input.
How it connects
Connects directed acyclic graphs, state machines, retrieval-augmented generation, vector databases, and large language models.
Cover of Investing for Programmers
liveBookmanning.com
Read Investing for Programmers on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Investing for Programmers. The section links point to the parts of the book it draws on.