Projects

docintelligence

An interface to chat with your pdfs (RAG) from google drive. Built with pgvector, django, react, openai API, postgreSQL, GCP, docker and github actions.

Tailored Mindfulness App

An AI-driven meditation app for personalized mindfulness sessions. Built with django, react, openai API, SQLite, GCP, docker and github actions.

Experiences

Freelance AI/ML Engineer

2025 - Present

AI/ML Engineer

2024 - 2025
Kantar Media

Full-Stack development.

Senior Data Scientist

2021 - 2024
Huawei, Quant AI Lab

LLMs, model evaluation and deployment, vector databases, Azure Devops.

PhD Student

2017 - 2020
Computer Vision Center, omni:us

Research on information extraction from semi-structured documents with neural networks.

Data Scientist

2015 - 2016
Hockerty, Ulabox

Data science internship, recommender systems, churn prediction.

Publications

  • Named Entity Recogntion and Relation Extraction with Graph Neural Networks in Semi-Structured Documents
  • Manuel Carbonell, Pau Riba, Mauricio Villegas, Alicia Fornés, Josep Lladós
    International Conference on Pattern Recognition, 2020 (oral)
  • A Neural Model for Text Localization, Transcription and Named Entity Recognition in Full Pages
  • Manuel Carbonell, Alicia Fornés, Mauricio Villegas, Josep Lladós.
    Pattern Recognition Letters, 2020
  • End-to-End Handwritten Text Detection and Transcription in Full Pages
  • Manuel Carbonell, Joan Mas Romeu, Mauricio Villegas, Alicia Fornés, Josep Lladós.
    International Conference on Document Analysis and Recognition Workshops, 2019
  • Joint Recognition of Handwritten Text and Named Entities with a Neural End-to-End Model
  • Manuel Carbonell, Mauricio Villegas, Alicia Fornés, Josep Lladós.
    International Conference on Document Analysis Systems, 2018

    Skills & Proficiency

    Python (pytorch, transformers, tensorflow, pandas, numpy, django, fastapi, pytest)

    Shell

    Azure DevOps, GCP

    SQL

    JavaScript

    C

    Java