Electrical and Computer Engineering - Master's degree 2018-2020
[IST] Instituto Superior Técnico, Portugal
Hi I'm Carlos Simões an enthusiastic engineer eager to learn and make a positive impact on the world through work and life. I strongly believe in technology as an enabler of change, and would like to help steer this change in the right direction.
My interests include:
This hard skills were developed through the projects, personal motivation, external courses and internships.
[IST] Instituto Superior Técnico, Portugal
University of Coimbra, Portugal
[ISEC] Instituto Superior de Engenharia de Coimbra, Portugal
High School Secondary Education at Avelar Brotero, Portugal
Nokia
- Developed and evaluated machine learning models and training data to improve performance and quality.
- Collaborated with cross-functional teams to understand requirements and implement them into technical specifications and features for use in chatbots.
- Improved the accuracy of the company's chatbot by 20\% through the use of machine learning and NLP techniques.
Accenture
Capable in Big Data technologies such as Hadoop and NoSQL.
Skilled in data analysis using Python, Pandas, numpy, Scikit-learn, and deep learning.
Experience with data visualization tools such as Excel, PowerBi, and Qlik Sense.
Familiar with ingestion and processing of large amounts of data on AWS.
Institute for Systems and Robotics (ISR-Lisboa)
For my thesis, I developed an algorithm that captures the Visual Focus of Attention from the gaze and head pose with the aim of improving Human-Robot Interaction. The algorithm used stereo systems to estimate depth and perform 3D reconstruction of the environment. We then used the 3D head and gaze pose to calculate and track the object that a person is interested in. To aid in environment classification, we employed panoptic segmentation to identify objects in the scene. The results of our work were published in a Paper. This project allowed us to investigate the use of stereo systems and 3D pose estimation for enhanced human-robot interaction and gaze tracking.
Núcleo de Estudantes de Engenharia Eletrotécnica e Computadores da Associação Académica de Coimbra, Portugal
Implementation of the "Delegates of the Year" project
Awareness and pedagogical training of teachers and students
Conducting pedagogical surveys and Pedagogical Forum
Meetings for discussion and curriculum reform of MIEEC
Direct and continuous contact with the DEEC community and management
Coimbra, Portugal
IJsselstein, Utrecht, Netherlands
Design and drawing in AutoCAD 3D and 2D for the BMV Veiden project. In charge of some additional work for other projects
For my thesis, I developed an algorithm that captures the Visual Focus of Attention from the gaze and head pose with the aim of improving Human-Robot Interaction. The algorithm used stereo systems to estimate depth and perform 3D reconstruction of the environment. We then used the 3D head and gaze pose to calculate and track the object that a person is interested in. To aid in environment classification, we employed panoptic segmentation to identify objects in the scene. The results of our work were published in a paper. This project allowed us to investigate the use of stereo systems and 3D pose estimation for enhanced human-robot interaction and gaze tracking.
Developed an end-to-end platform that uses machine learning to predict whether customers will or will not subscribe to a bank program. The project included exploratory data analysis, model building and evaluation, and visualization of the results. Using Python, Pandas, Matplotlib, Sklearn and PowerBI
Developed a gender recognition system using deep learning methods to scan eye images and determine the gender of the patient.} The system was able to accurately classify individuals as male or female. Using Python, Pandas, Numpy and Tensorflow
Real-time control of a smart office illumination system using Arduinos and Raspberry Pi: Developed a system for real-time control of an office's illumination using Arduinos and Raspberry Pi. The system was able to automatically adjust the lighting levels in the office based on real-time information, such as the time of day and the presence of people in the room. The use of Arduinos and Raspberry Pi allowed for a low-cost and flexible solution that could be easily integrated into existing office infrastructure. Made in C++, MATLAB and Arduino.
Project that consists of two other project on the subject of Artificial Intelligence, this project were both made in Python.
The travelling sales problem (TSP) and the ant colony optimization (ACO) algorithm: Developed and implemented the ant colony optimization (ACO) algorithm to solve the travelling sales problem (TSP). The TSP aims to find the shortest cycle in a weighted graph that passes through all its nodes, and the ACO algorithm uses a simulation of ants laying down pheromone trails to find solutions to the problem. The project was implemented in Java.
Interactive voice response system (IVR) based on voice over IP (VoIP): Developed an IVR system using the voice over IP (VoIP) protocol. The project involved understanding and implementing the relevant VoIP protocols, and the system was built using Python and config files. The IVR system allowed for the creation of automated voice menus and interactions with users via phone calls. .
Linear state feedback control of an inverted pendulum: Developed a control system for an inverted pendulum using linear state feedback. The project involved the design and implementation of the control algorithms in MATLAB, and the system was able to accurately control the pendulum's motion and maintain its stability. The use of linear state feedback allowed for a simple and effective control solution. .
Two Projects in the suject of Robotics, all made in MATLAB:
Transferring a robot from point A to B using multiple algorithms to estimate the optimal route: Developed algorithms to guide a robot from one point to another, taking into account the robot's capabilities and the characteristics of the environment. The project involved the use of Matlab to implement and evaluate the algorithms, and the resulting system was able to find efficient and safe routes for the robot to follow.
Veterinary hospital database and web application: Developed a database and web application for a veterinary hospital. The application allowed for the efficient storage, retrieval, and management of patient records and appointments. The web-based interface allowed for easy access to the database from any device with an internet connection. Using MySQL and PHP
Management of a car park using a graphic platform and database: Developed a system for managing a car park, including a graphic user interface and a database to store information about the cars and parking spaces. The project involved the use of C#, C/C++, and MSAccess, among other technologies, to implement the system's features. The system was able to accurately track the entry and exit of vehicles, and calculate the total amount to be paid based on the duration of the parking and any discounts.
Machine learning algorithms and libraries, such as TensorFlow, Keras, and scikit-learn.
Programming languages such as Python, matlab and C++
Data analysis and visualization tools, such as Pandas,Numpy and Matplotlib
Deep learning techniques, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), Transformers.
Computer vision techniques, including image and video processing, object detection, and face recognition and GANs.
Familiarity with Linux operating systems and shell scripting.
Portugal