AI 🤖 & Cibersecurity ⚔️

· GitHub · LinkedIn · TryHackMe · HackMyVM

Hi 👋, I am Hector! a master’s student in Cybersecurity and graduated in Computer Engineering (major in CS). I currently work researching on the security offered by different Coding techniques in the field of Federated Learning. I have a high background in Pentesting/Red-Team techniques and advanced knowledge of different Machine Learning techniques. Looking forward to continue learning and increasing my skills.


🥷 Skills


👨‍💻 Career Experience

GRADIANT

Sep. 2023 - Currently [Vigo & Remote]

Galician Telecommunications Technology Center. Innovation to improve the competitiveness of companies in the market.

  • State-of-the-art study of Coded Distributed Computing and Privacy Enharancing Techniques for Federated Learning.
  • Simulation of attacks and analysis of security flaws in a FL architecture in PySyft.
  • Implementation of a CDC scheme for secure aggregations inside PySyft’s core.

Torusware

Feb. 2022 - May. 2022 [Coruña & Remote]

Torusware helps organizations to outcompete through solutions in the areas of Big Data, High Performance Computing and microservices.

  • Designed an architecture to allow ML at the edge, in an hybrid-AWS cloud scenario.
  • Automation of different actions on the web using libraries such as Selenium, BeautifulSoup and HTTP requests in Python.
  • Migration from EFK Stack (Elasticsearch, Fluentd and Kibana) to a ELK Stack (Elasticsearch, Logstash and Kibana) in a Kubernetes cluster,since it was found that Logstash performed more efficient data ingest and processing for the scenario.

📚 Education

Masters Degree in Cybersecurity - Universidade da Coruña

2022 - 2024 [A Coruña]

Computer Engineering at University of A Coruña (CS major) - Universidade da Coruña

2018 - 2022 [A Coruña]

⚒️ Personal Projects

Network Intrusion Detection System

Evaluation of ML techniques for IDS

This project presents different techniques for the detection of anomalies in network flows. For this purpose, a labeled dataset, the CSE-CIC-IDS-2018, was used, on which different exploration and preprocessing phases were performed to facilitate model learning. The techniques used can be grouped according to the inspection methodology: Anomaly Detection (AD), Signature Detection (SD) and Hybrid Detection.

DR-Keylogger

Post exploitation tool to capture keystrokes ‘quietly’.

It is a post-exploitation tool to capture keystrokes from multiple end-users in a corporate network/LAN and send them to a central server. All the keystrokes are sent over HTTP and encrypted via a Double-Ratchet communication mechanism. The client-server communication scheme is designed to go unnoticed in high security networks.