Skip to content
View samuellimabraz's full-sized avatar
34 followers · 39 following

Achievements

Achievement: Pull Sharkx3Achievement: StarstruckAchievement: YOLOAchievement: Pair ExtraordinaireAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx3Achievement: StarstruckAchievement: YOLOAchievement: Pair ExtraordinaireAchievement: Quickdraw

Highlights

  • Pro

Organizations

@Black-Bee-Drones @Fundacao-Asimo

Block or report samuellimabraz

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
samuellimabraz/README.md

I am Samuel, a Robotics AI Engineer. I really like robotics, drones, computer vision and AI.
Horse
Portfolio

Current Projects & Work

  • Chunkr ML Engineer at Chunkr - Document layout analysis and computer vision

  • Bee      I'm part of Black Bee Drones - Autonomous drone competition, research team

Featured Projects

  • Nectar SDK Stars — ROS2 SDK for autonomous drones: unified interfaces for flight control, computer vision and object detection

  • Quantum Assistant Stars SSRN Preprint — Multimodal dataset and model for quantum computing and Qiskit. My final graduation project.

  • Signature Detection Model Stars — Open-source signature detection with YOLOv8, deployed with Triton (ONNX/OpenVINO). End-to-end computer vision project with more than 20 million downloads on Hugging Face.

  • Vision to MAVROS for ROS2 Stars — ROS2 bridge from Intel RealSense T265 pose to ArduPilot/PX4 via MAVROS, and utils for vSLAM with d435i and Isaac ROS.

  • PEFT: Fine-Tuning Methods for LLMs — Article and experiments studying and comparing main PEFT techniques for LLMs finetuning.

  • PID Controller for ROS2 Stars — General PID controller as a ROS2 node with reconfigurable parameters in real time

  • CafeDL Stars — Java deep learning library from scratch (Conv2D, Dense, optimizers, ND4J) plus a QuickDraw-style sketch game.

  • OpenCV Explorer — Web app for real-time computer vision experiments with the webcam (OpenCV, Streamlit, WebRTC).

  • Hand Mouse Controller — Mouse control with hand gestures in real time using MediaPipe.

GitHub Stats

Tech Stack & Tools

Languages Frameworks & Libraries Systems & Tools

Connect with Me

LinkedIn ORCID Kaggle Hugging Face Gmail

Pinned Loading

  1. Black-Bee-Drones/nectar-sdk Black-Bee-Drones/nectar-sdk Public

    ROS2 SDK designed to simplify drone control and computer vision tasks

    Python 6

  2. tech4ai/t4ai-signature-detect-server tech4ai/t4ai-signature-detect-server Public

    This project provides a pipeline for deploying and performing inference with the YOLOv8 object detection model using the Triton Inference Server. It supports integration with local systems, Docker-…

    Jupyter Notebook 22 4

  3. cafedl cafedl Public

    Design and implementation of a small DeepLearning library from scratch in Java, inspired by Keras and the book "Deep Learning From Scratch" by O'Reilly. The goal is to apply the main OOP Design Pat…

    Java 4

  4. Black-Bee-Drones/vision_to_mavros Black-Bee-Drones/vision_to_mavros Public

    Adaptation of the vision_to_mavros package for ros2 with realsense T265

    C++ 6

  5. quantum-assistant quantum-assistant Public

    Multimodal Dataset and Model for Quantum Computing and Qiskit

    Python 4

  6. Black-Bee-Drones/pid-controller Black-Bee-Drones/pid-controller Public

    Generic, configurable PID controller implemented as a ROS2 node in C++

    C++ 1