MSc Candidate in Computer Science @ UFMG — enthusiastic about AI, with a research track in vibration-based intelligent fault diagnosis and a current pivot into Music Information Retrieval (MIR) and music generation.
Profile: github.com/ivarejao

I work at the intersection of signal processing and machine learning:

• Past / published work: intelligent fault diagnosis for rotating machinery using vibration signals (incl. reproducible research code + benchmarks).
• Current focus: MIR with emphasis on music generation (bridging industrial time-series modeling ↔ creative audio/music modeling).

I’m interested in opportunities and collaborations in AI/ML, applied research, MIR, and music tech.

• Vibdata — a Python package providing a common interface to store/manipulate vibration datasets for ML in fault diagnosis
  Repo: ivarejao/vibdata
  Highlights:

  • Unified interface + multiple public datasets
  • Modules for raw dataset handling and deep (deep-learning-friendly, incl. PyTorch-oriented workflows)
  • Paper-reproducibility tags (see README)

• Similarity bias benchmarks (fault diagnosis) — benchmark code and experiments for the “similarity bias” problem in vibration-based ML
  Repo: ivarejao/vibdata-baselines
  Includes baselines like Nearest Neighbor, Random Forest, and a 1D CNN (XResNet18-1D), evaluated under multiple cross-validation strategies.

• “The similarity bias problem: what it is and how it impacts vibration-based intelligent fault diagnosis.”
  Code/benchmarks: ivarejao/vibdata-baselines
  DOI (as referenced in Vibdata): https://doi.org/10.1016/j.ymssp.2025.112822

• International Journal of Prognostics and Health Management paper (as referenced in Vibdata):
  DOI: https://doi.org/10.36001/ijphm.2025.v16i3.4239
  Reproducibility notes/tags: ivarejao/vibdata

• NLP-with-Disaster-Tweets — “Projeto Final de Data Science”
• veja-news-scraper
• T1-SO — “Vaccine shell” implemented in C (with co-authors)

• LinkedIn: https://www.linkedin.com/in/igor-varejao (from my GitHub profile: github.com/ivarejao)
• ORCID: https://orcid.org/0009-0002-3228-7141 (from my GitHub profile: github.com/ivarejao)
• X: https://x.com/igvarejao (from my GitHub profile: github.com/ivarejao)

• Research collaborations (AI/ML, robust evaluation, signal/audio modeling)
• Internships / applied ML roles
• MIR / music generation projects where strong evaluation + engineering matter

If any of that overlaps with what you’re building, feel free to reach out.