labrijisaad / llm-rag Goto Github PK

This image, generated with DALL-E, depicts a wide Moroccan landscape where ancient ruins and modern AI structures blend, symbolizing the harmony between the past and the future.

• 🌱 Hello, I'm Saad, a 23-year-old based in France, with a deep passion for creating projects in the realms of Data and Artificial Intelligence.
• 🎓 I hold a Data Engineering degree from INPT.
• 💼 Currently working as a Machine Learning Engineering Apprentice at AXA - Direct Assurance.
• 📚 I'm also preparing for a Master's degree in Machine Learning and Data Science at Paris Cité University.

Contributed to repackaging and updating the GIT Clustering algorithm 🔄 based on insights from an arXiv paper, with implementation available in the GitHub repository 📂 and distribution through the TestPyPI Package 📦.

• Machine Learning Engineer / Data Scientist Apprenticeship at AXA - Direct Assurance, Paris, France (Ongoing) More details
• Data Engineer / Data Scientist Internship at Chefclub, Paris, France (6 months) More details
• Data Engineer Intern at Capgemini Engineering, Casablanca, Morocco (2 months)
• Data Scientist Intern at AIOX Labs, Rabat, Morocco (2 months)
• Web/Backend Developer Intern at DXC Technologies, Rabat, Morocco (2 months)

• Repository: LLM RAG

Description: A Streamlit application leveraging a Retrieval-Augmented Generation (RAG) Language Model (LLM) 🤖 with FAISS indexing 🗃️ to provide answers from uploaded markdown files. Users can upload documents 📝, input queries, and receive contextually relevant answers using Similarity Search 🔍, showcasing a practical application of NLP technologies 🤖. The project is also equipped with a CI/CD pipeline 🔄 ensuring code quality & tests and simple deployment, and it supports containerization with Docker 🐳 for easy distribution and deployment.

• Technologies/Tools: Streamlit, OpenAI API Models (LLMs, Embedding Models), FAISS, Python, Docker, CI/CD (Github Actions), Makefile, venv.

• Repository: Kedro Energy Forecasting Machine Learning Pipeline

Description: A showcase of MLOps best practices using Kedro 🛠️, this repository shows the journey of Machine Learning Models from development to deployment 🚀, utilizing Docker 🐳. Featuring straightforward training, evaluation, and deployment of models such as XGBoost Regressor, LightGBM 💡 and Random Forest Regeressor 🌳, it integrates built-in visualization 📊 and logging 📝 for effective monitoring. Dive into the world of modular and scalable data pipelines with Kedro 📚 Kedro Documentation. The integration of an automated CI pipeline 🔄 with Github Actions ensures code quality ✅ and reliability 🔒.

• Technologies/Tools: Docker, Kedro, MLOps, CI/CD (Github Actions), Machine Learning (XGBoost, Random Forest, LightGBM), Jupyter Notebook, Makefile, venv, Python.

• Repository: GIT Clustering

Description: An upgraded version of the GIT Clustering algorithm 🔄, informed by insights from an arXiv paper 📄, with easy deployment via TestPyPI 📦. Includes benchmarking notebooks 📊 comparing it to state-of-the-art clustering algorithms 🔍.

• Technologies/Tools: Benchmarking, Poetry Packaging, PyPI Distributing, Machine Learning (K-means, DBSCAN, AgglomerativeClustering, Gaussian Mixture..), Jupyter Notebook, Makefile, venv, Python.

• Repository: Monthly & Daily Energy Forecasting Docker API

Description: This repository 📦 houses an Energy Forecasting API ⚡ that uses Machine Learning to predict daily 📅 and monthly 🗓 energy consumption from historical data 📊. It's designed as a practical demonstration of a ML Engeineering/Data Science workflow, from initial analysis to a deployable API packaged with Docker 🐳.

• Technologies/Tools: MLOps, Docker, API design, Machine Learning (XGBoost, Random Forest), Jupyter Notebook, Makefile, venv, Python.

Let's make something innovative together! Feel free to reach out for collaborations or discussions in Data & Artificial Intelligence!

• README last updated on 17/04/2024. Regularly updated to reflect current work and interests.