Juan Manuel
|
Elche, Alicante, Spain
Physicist and AI engineer specialized in computer vision. Solving complex problems with high-impact AI.
About me
Physicist and mathematician specialized in Machine Learning and Computer Vision engineering. I build computer vision systems that work in real production: from data capture and preprocessing to model deployment. I am drawn to the intersection between mathematical rigor and AI solutions with tangible impact. When I am not training models, I write about physics, chaos and meteorology on this blog.
...
Experience
0
Projects
AI
Specialist
Projects
Tire Defect Detection
Computer vision pipeline for automatic defect detection in the tire industry using deep learning.
AI Misinformation Detection
Ontology-based AI architecture for detecting misinformation in fiction works. Published at DCAI'25.
Lagrangian Atmospheric Blocking
Lagrangian method for automatic identification of atmospheric blocking situations using HYSPLIT trajectories.
Talk to Me
CLI for practicing conversational English with an AI tutor. Speak through your microphone, Whisper transcribes your voice in real time, and a local LLM (Ollama) keeps the conversation going.
Local Agent with Ollama
Conversational console agent running 100% locally on Ollama. Native tool calling: file read/write, web search with DuckDuckGo, Python code execution, and RAG with ChromaDB to query your own documents.
Frontend: An Introductory Course
Educational frontend web development repository. Covers HTML, CSS and modern layouts (Flexbox/Grid) with detailed theory and commented examples. Under progressive development.
Personal Portfolio
This website: a static portfolio built with Hugo and Tailwind CSS, deployed on GitHub Pages with CI/CD.
Experience
ML Engineer · Computer Vision & LLMs
I take defect detection and segmentation models (CNNs, ViT, YOLO, RF-DETR) and a multimodal OCR (GLM-OCR) to production for industrial tire inspection: mAP@50:95 of 0.83 and 98.56% exact match reading sidewalls.
I write about this work in OCR in extreme conditions and CNNs vs Vision Transformers.
Data Scientist & AI Researcher
Research in NLP and LLMs: an ontology-based misinformation-detection architecture built with LLM-generated ontologies (published at DCAI'25), RAG pipelines and multi-agent systems, plus computer vision projects.
Publication: DCAI'25 paper.
Researcher · Internship
I developed Lagrangian methods to automatically identify atmospheric blocking situations — the basis of my thesis and a publication in progress.
I tell the full story in Atmospheric blocking, a Lagrangian view.
Latest Blog Post
Education
Bachelor's Degree in Physics
Universidad de Alicante
+120 ECTS from the Mathematics degree taken in parallel (UA simultaneous-studies program).
Thesis: A Lagrangian method for the identification of atmospheric blocking situations
View ThesisSICUE Exchange — Physics & Mathematics
Universidad de Salamanca
International mobility scholarship.
Publications
A Lagrangian Method for the Identification of Atmospheric Blocking Situations.
AI-Powered Ontology-Based Architecture for Misinformation Detection in Fiction Works.
Download Paper