Jose Fuentes

About Me

I am a Ph.D. Candidate in Computer Science and Applied Mathematician at Florida International University, specializing in Computer Vision and Deep Learning for the Physical World. My research focuses on building perception systems that are mathematically rigorous, physically consistent, and deployment-ready. I develop frameworks leveraging Vision-Language Models, geometric vision, and probabilistic modeling for autonomous systems and environmental mapping. My work has been recognized with an ICRA 2025 Best Paper nomination and deployed on edge hardware including NVIDIA Jetson and Raspberry Pi platforms.

Research and Work Experience

Research Assistant

Florida International University

Miami, FL, United States

January 2022 - Present

Constructed AI algorithms for continuous surface and underwater monitoring, increasing long-term environmental surveillance and reducing human intervention at Biscayne Bay in Miami
Developed and enhanced autonomy for the SeaRobotics Surveyor ASV and BlueROV2 AUV by integrating perception sensors and implementing machine learning algorithms for environmental monitoring and surveillance
Implemented ML-based perception and navigation strategies to improve real-time decision-making and autonomy, enabling adaptive exploration and data-driven environmental management in aquatic environments
Led live demonstrations of multi-robot systems to stakeholders from key government and research agencies, including the Florida Department of Environmental Protection, Office of Naval Research, National Science Foundation, and the U.S. Army Corps of Engineers

Research Associate

Nueva Granada Military University

Hybrid / Bogotá D.C., Colombia

December 2021 - August 2022

Created a Python library for processing hexagonally sampled images, addressing limitations in the angular resolution of classic image grids and improving spatial processing
Implemented a visual attention model in Python for hexagonally sampled images, inspired by human eye cells: cones and rods, increasing focus detection and image recognition

Student Lecturer

National University of Colombia

Bogotá D.C., Colombia

August 2018 - December 2020

Taught Linear Algebra and Calculus I, II, and III as a main instructor, participating in study groups to develop new teaching strategies for undergraduate students and providing personalized assistance to improve student performance
Provided academic support in mathematics classes, including personalized help to students, boosting understanding and performance through targeted assistance and innovative teaching approaches

Research Consultant

National University of Colombia

Bogotá D.C., Colombia

November 2020 - February 2021

Developed a dynamic crime prediction model for cities divided into quadrants, collaborating with local police in several major cities to enhance spatial accuracy and urban safety strategies
Documented and implemented predictive models based on spatial diffusion and state-of-the-art crime models in Python
Cooperated within a group of teams focused on predicting, optimizing, and enhancing police response across multiple cities, collaborating closely with local authorities to improve urban safety measures

Research Assistant - Invited Member

National University of Colombia Cim@lab Research Group

Bogotá D.C., Colombia

January 2020 - August 2021

Researched in the extraction of key features from medical imaging modalities, including MRI and procedural videos, to inform machine learning applications
Developed automated machine-learning techniques to enhance super-resolution imaging
Facilitated photo documentation in gastrointestinal procedures to improve image clarity and diagnostic utility

Education

January 2022 - July 2026

Ph.D. in Computer Science

Florida International University, Miami, FL

Specializing in Computer Vision and Deep Learning. GPA: 4.0. Research focus on multimodal perception, geometric vision, and probabilistic modeling for autonomous systems. Nominated for ICRA 2025 Best Paper Award. Coursework: Analysis of Algorithms, Operating Systems, Artificial Intelligence, Machine Learning.

August 2018 - April 2021

M.S. in Applied Mathematics

National University of Colombia, Bogotá D.C.

GPA: 3.7. Received meritorious distinction for master thesis on "Restoration of blurred images using linear programming." Coursework: Convex Optimization, Numerical Analysis, Image and Signal Processing, Partial Differential Equations.

January 2013 - December 2017

B.S. in Mathematics

National University of Colombia, Bogotá D.C.

Recognized as one of the best students in the sciences department senior cohort (Efi-Ciencias Exam Winner, Mathematics Section, 2017). Received Teaching Assistant Scholarship (2018-2021) to teach while studying.

Projects

Surveyor Gradient Tracking

ICRA 2025 finalist implementation: Learning-based adaptive navigation algorithm for scalar field mapping on Autonomous Surface Vehicles (ASVs). Nominated for Best Paper, Best Student Paper, and Best Paper in Field and Service Robotics.

PythonRaspberry PiSci-kit LearnSeaRobotics Surveyor ASV

Live → GitHub →

SeaRobotics Surveyor Control

Control library for ASV SeaRobotics platform with comprehensive sensor integration, IMU handling, and high-level autonomous control capabilities for environmental monitoring missions.

PythonRaspberry PiYSI EXO2Real-time Systems

Live → GitHub →

BlueROV Control Library

Python-based control interface for BlueRobotics underwater vehicle with comprehensive sensor suite and actuation support for marine robotics research.

PythonRaspberry PiOpenCVUnderwater RoboticsSlamtec RPLIDAR

Live → GitHub →

TidalFlow-SWE

PDE solver for shallow water equations with applications to tidal simulation and environmental modeling. Implements high-performance numerical methods for fluid dynamics.

PyClawPythonMPIPETScPandasNumPy

Live → GitHub →

PhysRAG-SWE

Physics-informed framework for shallow water equation modeling and analysis. Combines retrieval augmented generation with domain expertise for improved predictions.

PyClawPythonMPIPETScPandasNumPy

Live → GitHub →

Advection-Diffusion Simulator

FEniCS-based numerical solver for advection-diffusion processes. Demonstrates high-performance PDE computation with applications to environmental transport modeling.

FEniCSPythonMPIPETSc

Live → GitHub →

Hexagonal Image Processing Library

High-performance C++/Python library for processing hexagonally sampled images. Improves angular resolution and feature extraction on non-standard sensor grids.

NumPyPythonOpenCV

Live → GitHub →

C++ Graph & Monte Carlo Library

High-performance implementations of graph algorithms and Monte Carlo simulations with applications to hexagonal grid processing and probabilistic modeling.

C++

Live → GitHub →

PawPal

Streamlit-based pet owner activity scheduler demonstrating full-stack web application development. Features activity scheduling, pet profile management, and user authentication.

StreamlitPythonNumPyPandas

Live → GitHub →

Research

Autonomous Robotic Information Gathering Systems

Xioeng

Learning-based Adaptive Navigation for Scalar Fields

Jose Fuentes

Localization in Sensory-Compromised Environments using Environmental PDEs

Jose Fuentes

Towards Optimal Human-Robot Interface Design for Underwater Teleoperation

Jose Fuentes

Skills

Programming Languages and Frameworks

Python

0 %

C++

0 %

MATLAB

0 %

PyTorch

0 %

TensorFlow

0 %

Skills & Tools

Python (Proficient)

MATLAB (Advanced)

C++

Bash

TensorFlow

Scikit-Learn

ROS

Pandas

NumPy

OpenCV

Optimization

Linear Algebra

Partial Differential Equations

Data Analysis

Problem Solving

Computer Sciences

FEniCS

Git

Docker

Linux/Unix

LaTeX

Raspberry Pi

Arduino

NVIDIA Jetson

Spanish (Native)

English (Fluent)

German & Portuguese (Beginner)