COUTO PINTOS MANUEL

COUTO PINTOS MANUEL

@manuelcoutopintos

Universidade de Santiago de Compostela En ITH desde 6/8/2026
rustsveltepytorchnlpsupabasepostgresscloudflare

Manuel Couto Pintos

manucouto1

Software Engineer & AI/NLP Researcher

AI PhD Candidate at CiTIUS (USC), specializing in Natural Language Processing. I merge my research background with full-stack software development, from robust backend systems to interactive web interfaces.


🛠️ Tech Stack & Tools


🚀 Featured Projects

A Tool for Building and Evaluating Customizable LLM Assistants.

Next.jsTypeScriptVercel

Visit Website →

A framework for building, experimenting with, and evaluating LLM chains (agents) in a systematic and reproducible way.

PythonLangChainModularML

Visit Website →

Research on the effectiveness of different embedding models (Word2Vec, GloVe, BERT) for measuring semantic coherence in Topic Modeling.

PythonGensimPyTorchspaCy

Early detection of mental health risks using temporal dynamics of word embeddings on social media. Published in JHIR (Q1).

PythonGensimPyTorchBERT

Read Paper →

🧠 Hierarchical Attention Pooling

Hierarchical model for early mental health detection. Builds chunk-level temporal representations via multi-level attention.

PythonPyTorchAttentionLSTM

Repository private for now

🔬 New Research in Progress

Currently finalizing new projects on applied AI, autonomous agents, and P2P systems. Stay tuned!

AILLMsRustP2P

📝 Recent Publications

  • Temporal Word Embeddings for Early Detection of Psychological Disorders on Social Media: Published in Journal of Healthcare Informatics Research (Q1 JCR). Read Paper
  • Comparison of Clustering Algorithms for Knowledge Discovery in Social Media Publications: A Case Study of Mental Health Analysis: Presented at SEPLN 2024. Read Paper

Repos de GitHub

Proxectos

Colaboracións

Aínda sen colaboracións rexistradas.