Skip to content

Each project demonstrates a unique application of Large Language Models (LLMs), combining them with web scraping, custom tool calling, code translation, and retrieval-augmented generation (RAG) techniques. All projects feature interactive Gradio frontends, enabling real-time, user-friendly interaction with powerful AI backends.

Notifications You must be signed in to change notification settings

SionAlin/LLM_Engineering

Repository files navigation

πŸš€ AI Projects Portfolio

This repository contains 4 AI-powered projects, each demonstrating different capabilities of Large Language Models (LLMs) and integrations with tools, APIs, and UIs.
All projects use Gradio for interactive frontends and integrate with OpenAI / Ollama / LangChain as needed.


1️⃣ Proj1 β€” 🌐 AI Brochure Generator

Description:
Scrapes a company website, intelligently selects relevant pages (About, Careers, etc.), and generates a beautiful brochure in Markdown in any language.

Key Features:

  • 🌍 Web scraping with BeautifulSoup
  • πŸ€– OpenAI GPT models & Ollama integration
  • πŸ“œ JSON link filtering using LLM reasoning
  • 🎨 Gradio UI for easy interaction

Screenshot:
Proj1 Screenshot


2️⃣ Proj2 β€” ✈️ FlightAI Assistant

Description:
A conversational assistant for an airline that:

  • Answers travel questions ✈️
  • Generates city images 🎨
  • Checks ticket prices πŸ’°
  • Shows ticket availability 🎟️

Key Features:

  • πŸ› οΈ Custom function calling tools (price, image, availability)
  • πŸ–ΌοΈ DALLE-3 city illustrations
  • πŸ’¬ Context-aware short replies
  • πŸ“± Gradio chat interface with image output

Screenshot:
Proj2 Screenshot


3️⃣ Proj3 β€” πŸ’» Python/Cobol β†’ C++ High-Performance Converter

Description:
Converts Python or COBOL code into optimized C++ for maximum speed.
Includes execution capabilities for both original and converted code.

Key Features:

  • πŸ”„ GPT-powered code translation
  • ⚑ High-performance C++ generation with -Ofast
  • ▢️ Built-in execution for Python, COBOL, and C++
  • πŸ–₯️ Gradio UI for conversion & execution

Screenshot:
Proj3 Screenshot


4️⃣ Proj4 β€” πŸ“š RAG AI Assistant with Chroma & LangChain

Description:
A Retrieval-Augmented Generation (RAG) assistant that answers questions using documents from a knowledge base.
Includes 2D & 3D vector visualizations of embeddings.

Key Features:

  • 🧠 LangChain + Chroma vector DB integration
  • πŸ“Š Plotly visualizations of embeddings (TSNE)
  • πŸ’¬ Conversational memory for context retention
  • πŸ–₯️ Gradio chat interface

Screenshot:
Proj4 Screenshot


πŸ› οΈ Tech Stack

  • Language: Python 🐍
  • Frameworks & Libraries:
    • Gradio, LangChain, BeautifulSoup
    • OpenAI API, Ollama, ChromaDB
    • Plotly, Matplotlib, Scikit-learn

πŸŽ“ Credits

These projects were developed as part of the course:
LLM Engineering: Master AI, Large Language Models & Agents
by Ed Donner


πŸ“Œ How to Run

  1. Clone the repo
git clone https://github.com/SionAlin/LLM_Engineering.git
  1. Add your OpenAI API key to .env
OPENAI_API_KEY=sk-yourkeyhere
  1. Navigate to the desired project folder and run:
jupyter notebook Main.ipynb

or

python Main.py

About

Each project demonstrates a unique application of Large Language Models (LLMs), combining them with web scraping, custom tool calling, code translation, and retrieval-augmented generation (RAG) techniques. All projects feature interactive Gradio frontends, enabling real-time, user-friendly interaction with powerful AI backends.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published