Systems Engineering student building practical systems where software, security, and AI meet.
I am a Systems Engineering student with a solid foundation in programming tools, AI and cybersecurity, specializing in Web Development with a strong interest in information security and data analysis. My goal is to grow as a developer and contribute to secure, innovative technological solutions — combining technical expertise with creative problem-solving.
I work across software engineering, AI/ML and full-stack development with a product engineering mindset: research-driven, security-conscious, and focused on shipping systems that work in practice. AI supports my workflow — research, prototyping, review — while architecture, security and final decisions stay human.
Open To: software engineering internships · cybersecurity & AI research collaborations · open-source collaboration.
Languages
Frontend
Backend & Databases
Cloud, DevOps & Tooling
Working proficiency — honest and conservative. Five-point scale; three is the current ceiling. Nothing here means expert.
1 minimal — initial contact · 2 low — small tasks · 3 medium — practical project experience
| Domain | Proficiency | Details |
|---|---|---|
| Machine Learning & Deep Learning | Intermediate | PyTorch, scikit-learn; deep neural networks trained on EMBER and SOREL-20M for malware detection |
| Applied AI for Cybersecurity | Intermediate | Hybrid ML + heuristic detection pipelines; static PE analysis at scale |
| Model Deployment & Inference | Intermediate | ONNX Runtime production inference; integration of trained models into real systems |
| Generative AI & LLM Integration | Familiar | Local models with Ollama; LLM-assisted explanations; OpenAI and Anthropic APIs |
| Data Science & Analysis | Familiar | NumPy, SHAP; exploratory analysis and feature engineering |
| AI-Assisted Development | Familiar | AI-accelerated research, prototyping and code review with Cursor, ChatGPT and Claude |
🛡️ ShadowNet Defender — intelligent malware detection with deep learning
Static malware detection for Windows PE binaries — a 2,381-d feature pipeline, DNN/ONNX inference, layered heuristics (YARA, PE, .NET, overlays) and local LLM explanations via Ollama.
| Stack | Python · PyTorch · ONNX Runtime · YARA · Ollama · Java |
| Scale | 2,381-dimensional feature pipeline; trained on large-scale datasets (EMBER, SOREL-20M) |
| Performance | 98.15% accuracy · 1.2% false positives · 0.65% false negatives (published IEEE results) |
| Security | Layered static analysis — PE parsing, YARA rules, .NET inspection, overlay detection — with human-readable local LLM explanations |
| Impact | Published at IEEE Xplore (Doc. 11214618) — intelligent malware detection system |
| Repository | https://github.com/IVAINX18/Shadow-Net-Defender-Hybrid-Malware-Detection |
A research-grade detection system that combines deep learning with classical static analysis: the neural model handles large-scale classification while layered heuristics catch structural anomalies, and every verdict is explained in plain language by a local LLM — no cloud dependency, no data leaving the machine.
🔒 Secure Call — real-time secure video calling
Real-time video-calling application built with React, Vite, PeerJS and WebRTC; secure peer-to-peer audio/video through a lightweight client-server architecture with a dedicated Express + Socket.IO signaling server.
| Stack | React · Vite · PeerJS · WebRTC · Express · Socket.IO · Tailwind CSS |
| Scale | Peer-to-peer media — streams flow directly between peers; server only for signaling |
| Performance | Direct P2P audio/video with minimal server load |
| Security | Standards-based WebRTC media encryption; isolated signaling layer |
| Impact | End-to-end secure communication tool shipped and documented |
| Repository | https://github.com/IVAINX18/Secure-Call |
A full-stack real-time communication project demonstrating modern frontend engineering and network architecture: the client handles media and UI while a minimal signaling server coordinates sessions — keeping the media path direct, private and efficient.
Hotel Website — multi-page hotel site, TalentoTECH capstone
Multi-page hotel website built for the TalentoTECH bootcamp — landing, booking, reservation management and inventory pages with gallery and responsive layout.
| Stack | HTML · CSS · JavaScript · PHP |
| Scale | Multi-page site (landing, booking, reservations, inventory) |
| Performance | Responsive layout across devices |
| Security | Baseline input handling and standard web practices |
| Impact | Bootcamp capstone project |
| Repository | https://github.com/IVAINX18/Pagina-de-Hotel |
A complete bootcamp capstone covering the full lifecycle of a small web product: multiple content pages, booking flows, reservation and inventory management, and a responsive gallery — from markup to deployment.
BMI Calculator — interactive health utility
Interactive Body Mass Index calculator with input validation, visual feedback and health recommendations.
| Stack | HTML · CSS · JavaScript |
| Scale | Single-page interactive utility |
| Performance | Instant client-side computation, zero dependencies |
| Security | Client-side input validation |
| Impact | Early foundations project — clean UX fundamentals |
| Repository | https://github.com/IVAINX18/Calculadora_IMC |
A focused foundations project built around one interaction done well: validated inputs, instant feedback and clear health guidance — demonstrating clean vanilla JavaScript and UX fundamentals without any framework overhead.
Systems Engineering Student — Universidad Cooperativa de Colombia · Feb 2022 – Present
Undergraduate program covering software engineering, networks, databases and information systems.
- Coursework across software development, networking fundamentals and database design
- Applied academic foundations in research and personal projects
Java SQL Networking Software Engineering Databases
Independent Developer & Researcher — Self-directed · 2024 – Present
Research-driven projects across software engineering, cybersecurity and applied AI.
- Built ShadowNet Defender, published at IEEE Xplore
- Developed Secure Call, a real-time P2P video-calling application
- Maintains a live portfolio documenting selected work and research
Python PyTorch ONNX React Security Research
| Recognition | Details |
|---|---|
| IEEE Xplore Publication | "ShadowNet Defender: Development of an Intelligent System for Malware Detection Based on Deep Learning" (2025) — https://ieeexplore.ieee.org/document/11214618 |
| ORCID iD | 0009-0000-3300-3724 — research identity |
| GitHub Pull Shark ×2 | Multiple merged pull requests |
| TalentoTECH Bootcamp | Completed programming bootcamp (2024) |
Learning — 
Building — 
Exploring — 
Open To — 
"Security is a process, not a product." — Bruce Schneier