Machine Learning & AI Engineer in training
Computer Engineering · Reinforcement Learning · Computer Vision · TinyML · AIoT
I build applied AI systems that connect models to real devices, software and operational problems. My work spans humanoid-robot decision policies, embedded computer vision, multimodal sensor fusion, data analysis and AI infrastructure.
I am a Computer Engineering student at Universidad de La Sabana with an IoT concentration and graduate-level coterminal coursework in Artificial Intelligence. I also work on AI/software solutions and contribute to reinforcement-learning research for the Sabana Herons humanoid robotics team.
| Project | My contribution | Core stack |
|---|---|---|
| Sabana Herons 2026 · team collaboration | Integrated role-conditioned RL policies into the C++ robot stack through ONNX; worked on the observation bridge, SimRobot scenarios, ball perception and HSL match behavior deployed for RoboCup 2026 | C++, Python, PPO, ONNX, SimRobot, YOLO, humanoid robotics |
| AIoT Posture & Fatigue Monitor · team lead | Developed two TinyML vision models and fused their output with MPU6050 inertial signals on an ESP32-S3; reported 199-222 ms embedded classification latency | TinyML, MobileNetV2, ESP32-S3, Edge Impulse, MQTT |
| Construction Safety PPE Detection · collaboration | Added the YOLO object-detection workflow, Colab execution and model-sharing documentation for detecting PPE compliance and visible violations | Python, YOLOv8, computer vision, Gradio, Jupyter |
| Paddle Tournament Predictor · collaboration | Contributed to the EDA, feature pipeline, logistic-regression ranking and functional match/tournament simulation exposed through Streamlit | Python, scikit-learn, Streamlit, feature engineering |
| Resil-IA · team collaboration | Co-developed a geospatial flood-risk prototype with an interactive map, community features and an n8n RAG assistant | React, Node.js, Leaflet, GeoJSON, n8n |
| Connect4RL · collaboration | Built the initial Connect Four environment, policy interface, dynamic policy discovery and tournament runner; the project evolved with Fermin Escalona's MCTS variants, timing controls and final policy | Python, MCTS, UCB1, Monte Carlo methods |
- Machine learning systems and applied AI engineering
- Reinforcement learning and decision-making under constraints
- Computer vision, TinyML and multimodal sensor fusion
- Robotics, AIoT and edge deployment
- Reproducible experimentation and reliable model evaluation
AI & data
Deployment & systems
I am focused on Machine Learning Engineering, AI Engineering and applied AI research internships. I am especially interested in teams where I can work across experimentation, implementation and deployment.
Languages: Spanish (native), English (B2 / IELTS 6.5), French (basic).