I'm the problem, it's m… ehm, that wasn't it. Okay, let's start again.
Hi! I'm Manuel 😁
A Junior Data Scientist who started out studying literature, linguistics, and storytelling. These days, I build tools to explore how language works using Python, NLP, and AI.
My background in Philology and Digital Humanities taught me to look closely at words. That same attention to detail now guides my approach to data science and machine learning.
I focus on natural language processing, data extraction, and text analytics, especially in cultural and creative domains.
I'm constantly learning, love iterative design, and care about how users actually interact with data tools.
I don’t just crunch text.
I try to listen to it.
| Project | Focus | Tools & Stack | Status |
|---|---|---|---|
| Lingua Animae | NLP classification (Bible) | Python, HF Transformers, Streamlit | MVP |
| Scrape-The-Verse | ETL + Literary Analytics | Spotify API, Genius, Power BI, NLP libs | Production |
| Resume Optimization | LLM-based resume rewriting | OpenAI/Gemini, Markdown, PDF export | In Development |
A thematic and emotional classifier for Bible verses using NLP and open-source models.
This MVP project classifies user input by emotion and theme, and connects it to relevant Bible verses.
It combines web scraping, manual annotation, Hugging Face models, and an interactive frontend using Streamlit.
💬 Over 80% of beta testers described the experience of literary empathy as meaningful and necessary.
An ETL pipeline that combines Spotify metadata and song lyrics to explore literary value in music.
This project extracts and transforms Spotify data and lyrics for artists like Bob Dylan and Taylor Swift. It includes metadata scraping, lyric alignment, and exploratory analysis of lyrical complexity and literary quality. Created interactive dashboards in Power BI to explore and compare the lyrical complexity of Bob Dylan and Taylor Swift.
An LLM-powered pipeline that adapts resumes to job descriptions and exports them to PDF.
This project streamlines the process of resume customization using LLMs (GPT-based or Gemini).
It converts .docx files into structured Markdown, adapts the content to match specific job offers, and exports polished, reader-friendly PDFs.
The pipeline includes keyword alignment, language optimization, and fully modular Python scripts ready for reuse and iteration.
I'm always open to new opportunities or collaborations.
Feel free to reach out!
