Welcome to my GitHub profile! I'm Jacques Gariépy, a passionate software engineer and researcher focused on creating innovative solutions that combine cutting-edge technology, traditionnal, AI-driven systems, agentic AI, and advanced paradigms to push the boundaries of software development.

• 🤓 I love discovering the latest trends! ✨ Learning new programming languages, exploring innovative frameworks, it really motivates me, creating. 💪 As a software engineer, I find it essential to stay up to date and always look for improvement.
• 🧠 A researcher and developer focused on transforming modern programming paradigms to align with AI-driven methodologies and deep learning innovations.
• 🔭 Passionate about integrating C#, Python, Angular, LLMs, and AI agents into scalable and innovative solutions.
• 🌱 Exploring new techniques in neural networks, modular AI systems, prompt engineering, and agentic intelligence to enhance user experiences.
• 🎨 Dedicated to creating tools that simplify complex workflows for developers while leveraging technologies like RAG, LLMs, and advanced model optimization.
• 💬 Ask me about dev, web, agentic ai, deep learning, neural networks, AI models, or anything related to cutting-edge software engineering and research.

In addition to my public repositories, I am actively working on numerous private projects, including advanced LLM models, cutting-edge AI systems, and many more innovative solutions designed to push the boundaries of technology and creativity. Stay tuned for future updates as these projects evolve!

Windows automation system powered by a Large Language Model (LLM) to understand and execute instructions. The idea is simple: you give a natural language command (e.g., “Open Notepad and type ‘Hello, world!’”), and the system attempts to translate it into actual actions on your Windows machine.

• ✨ Main features:

  • LLM-Driven Action Generation: The system interprets requests and dynamically generates Python code to interact with applications.
  • Automated Windows Interaction: Opening and controlling applications using tools like pywinauto and pyautogui.
  • Screen Analysis & OCR: Capture and analyze the screen with Tesseract OCR to verify UI states and adapt accordingly.
  • Speech Recognition & Text-to-Speech: Control the computer with voice commands and receive spoken feedback.

Out-of-code Insights is a Visual Studio Code extension that allows you to add annotations, comments, and notes without modifying your source files. It is ideal for avoiding clutter in your code with temporary comments or making the code unnecessarily heavy.

https://marketplace.visualstudio.com/items?itemName=JacquesGariepy.out-of-code-insights

Try Out-of-code Insights today and streamline your workflow without overloading your code!

• ✨ Main features:
  • Non-intrusive annotations: Add comments without altering the source code, keeping your files clean and organized.
  • Optimized for code reviews: Facilitate team communication by adding comments directly on the relevant lines.
  • Avoid code clutter: Maintain clarity and readability of your code by avoiding unnecessary comments.
  • Annotations on all file types: Add annotations to any file in your project, including source code, Markdown, JSON, XML, text, etc.
  • Intelligent change tracking: Annotations automatically follow file changes as long as edits are made within Visual Studio Code.
  • Personalization: Adapt the extension to your needs with various configuration options.
  • Threaded discussions: Reply to annotations to create threads and enhance collaboration.
  • Advanced annotation management: Activities bar, Move, filter, sort, and navigate annotations with ease.

This script is designed to analyze Python libraries, extracting detailed information about their elements, such as classes, methods, functions, properties, and more. It can save the analysis results to a JSON file for further inspection.

• ✨ Main features:
  • Detailed analysis of Python library elements
  • JSON export of analysis results
  • High Python code composition (98.1%) with minimal Dockerfile usage (1.9%)
  • Support for semantic search using BERT and whoosh

A cutting-edge language model that blends hyperbolic geometry with multi-head attention to capture complex, hierarchical structures in language.

• ✨ Main features:
  • Advanced modeling of linguistic structures
  • Integration of geometric concepts into language models
  • Potential for improved performance on complex tasks

Proof of concept to perform benchmarks on different methods using BenchmarkDotNet.

• ✨ Main features:
  • Benchmarking methods to evaluate performance
  • Use of BenchmarkDotNet for precise measurements
  • Comparative analysis of different implementations

A project leveraging AutoGen for creating autonomous agents with Tree of Thought (ToT) reasoning capabilities.

• ✨ Main features:
  • Advanced reasoning using the Tree of Thought framework
  • Integration of AI agents with AutoGen
  • Focused on modular AI workflows and prompt engineering

A dynamic system designed for optimizing neural network operations and AI workflows.

• ✨ Main features:
  • Advanced inference mechanisms for AI models
  • Optimization for real-time applications
  • Modular and scalable architecture

A project integrating AI with real-time weather data from OpenWeatherMap.

• ✨ Main features:
  • Agent-based weather forecasting
  • AutoGen-enhanced integration
  • Real-time API data processing

A robust system for handling service failures to ensure application continuity.

• ✨ Main features:
  • Configurable retry policies
  • Circuit breaker for service health monitoring
  • Rate limiting to prevent overload

An AI-powered command-line tool for analyzing, refactoring, and auditing source code.

• ✨ Main features:
  • In-depth code analysis
  • Intelligent refactoring suggestions
  • Automatic documentation generation

An experimental project aimed at making GPT-4 fully autonomous.

• ✨ Main features:
  • Implementation of autonomous GPT-4 workflows
  • Python-based experiments
  • Insights into next-gen GPT capabilities

A tool for simulating and visualizing swarm intelligence behaviors in various scenarios.

• ✨ Main features:
  • Simulation of multi-agent behaviors
  • Visualization of swarm dynamics
  • Configurable environments for testing

A collaborative chat application built with React and TypeScript.

• ✨ Main features:
  • Real-time communication for multiple users
  • Intuitive and modern UI
  • Lightweight and scalable architecture

A demonstration of the Observer design pattern implemented in ASP.NET MVC.

• ✨ Main features:
  • Practical implementation of the Observer pattern
  • ASP.NET MVC-based architecture
  • Simple and easy-to-follow example

A project aimed at simplifying metadata tagging and management.

• ✨ Main features:
  • Easy-to-use tagging interface
  • Integration with external systems for metadata organization
  • Lightweight and efficient tagging engine

A project aimed at creating custom GPT models for specific applications.

• ✨ Main features:
  • GPT models tailored to particular domains
  • Tools for model development and deployment
  • Detailed documentation for users

A Python application using AI to generate personalized recipes based on available ingredients.

• ✨ Main features:
  • Personalized recipe generation
  • Integration with GPT-4 models
  • User-friendly command-line interface

A PowerShell script for automating software installations and configurations.

• ✨ Main features:
  • Simplifies the software installation process
  • Automation of configurations
  • PowerShell-based for cross-environment compatibility

A cheat sheet for common Git commands and their uses.

• ✨ Main features:
  • Comprehensive list of Git commands
  • Usage examples and practical tips
  • Beginner-friendly reference

A project exploring the use of powerful language models for various applications.

• ✨ Main features:
  • Integration of advanced language models
  • Applications in natural language processing
  • Documentation and usage examples

A project aimed at predicting student performance and providing educational interventions based on influential factors.

• ✨ Main features:
  • Prediction model based on RandomForestRegressor
  • Analysis of feature importance affecting performance
  • Personalized educational recommendations

A collaborative project exploring lightweight multi-agent orchestration, managed by the OpenAI solutions team.

• ✨ Main features:
  • Educational framework for multi-agent orchestration
  • Ergonomic and lightweight agent management
  • Examples and use cases

A repository dedicated to prompt engineering, containing examples and guides to optimize interactions with language models.

• ✨ Main features:
  • Examples of prompts for various applications
  • Guides on creating and optimizing prompts
  • Resources for developers working with language models

A repository containing information and resources on deep learning.

• ✨ Main features:
  • Articles and tutorials on deep learning
  • Resources for beginners and experts
  • Links to online courses and recommended books

You can find me on these platforms:

This project is licensed under the MIT License ;)

Thank you for visiting my profile! Feel free to explore my projects and reach out if you'd like to collaborate or just chat.