A system brought to life

Vladimir Sibirov on building software systems and engineering teams

AI Engineering Goes Visual: Building an LLM RAG with PyFlyde & LangChain

Use Flyde to design and wire up LangChain, Vector Store, and Ollama or OpenAI API to perform Retrieval Augmented Generation

This part builds upon the Scraper app that we created in the part 1 and uses our visual programming skills to dive deeper in the LLM engineering world. This time we turn our article database into a RAG app, making use of LangChain, Vector Store and local LLaMA3 or OpenAI API.

AI Engineering Goes Visual: Web Scraping & Data Prep with PyFlyde

Learn how to create a web scraper and local article database using PyFlyde, BeautifulSoup4, and Playwright

In this tutorial, we explore how to use Flyde, a visual programming tool, to build a web scraper that feeds into a Retrieval Augmentation Generation (RAG) system. We will cover the process of scraping web content and storing it locally, setting the stage for more advanced AI engineering tasks.

The state of Flow-based Programming

Where Flow-based Programming stands after 50 years since its introduction. What problems it solves today. And what problems it can solve tomorrow for broader Software Engineering industry

Forcing a specific NoFlo version for all dependencies

Using NPM Shrinkwrap to enforce the same version of NoFlo for all dependencies to fix backwards compatibility issues

Development efficiency challenges in NoFlo

Obstacles on the way to happy day-to-day life with NoFlo and ways to overcome them

On NoFlo database applications

A few thoughts on design of NoFlo backends frequently talking to a RDBMS

Running web services as NoFlo graphs

Some problems that occur when trying to run entire web services as NoFlo graphs and some (radical) ways to solve them

Comparison of dataflow models

Comparison of Flow-based programming, Actor model and Communicating Sequential Processes in a scientific manner