Industry Experience

Professional Experience

Data Scientist — Faculty of Medicine, University of Chile
Neurosystems, SCIANLab & AudioBrain (2019–2025)

• Led interdisciplinary AI projects in healthcare, working closely with clinicians and domain experts
• Designed and deployed machine learning and computer vision systems for diagnostics, biomedical image analysis, and cognitive health assessment
• Applied signal processing and statistical modeling to physiological and clinical data under real-world constraints

Applied Research Engineer — University of Chile
(2009–2018)

• Designed and implemented computer vision and automation solutions for mining, retail, and biomedical applications
• Contributed to industry-funded R&D projects, supporting real-world system deployment and technology transfer

Selected Projects and Achievements

• Designed a machine learning pipeline for automated glaucoma diagnosis using chromatic pupilometry
• Applied spectral analysis and ML to EEG data for postoperative risk prediction
• Developed RaViTT, a Vision Transformer-based model for biomedical image classification
• Built deep learning models for 3D nuclei segmentation, achieving 2nd place in the Cell Tracking Challenge
• Co-developed VolumePeeler, a FIJI plugin for 3D image visualization and quantification of complex biological datasets
• Designed audiological tests for early detection of cognitive decline using signal processing and machine learning
• Contributed to the development of non-invasive biomarkers for early cognitive decline based on cochlear function
• Co-authored large-scale studies on social isolation and loneliness in older adults, analyzing post-pandemic mental health and demographic differences

Applied AI in Healthcare

• Machine learning systems for clinical decision support and biomedical data analysis
• Medical image analysis using deep learning (MRI, mammography, microscopy)
• Integration of signal processing with AI for physiological data (EEG, pupilometry, otoacoustic emissions)
• Development of reproducible and scalable pipelines for clinical and research environments

Computer Vision Systems

• Object detection, segmentation, and classification in real-world environments
• Feature extraction and pattern recognition for complex visual data
• Vision-based systems for industrial, mining, and biomedical applications
• Multi-scale texture and color analysis for automated lithological and microscopy image classification

Data Science & Machine Learning

• End-to-end ML pipelines:
  • data acquisition and preprocessing
  • feature engineering and selection
  • model training, validation, and evaluation
  • deployment, monitoring, and reproducibility
• Experience with:
  • supervised and unsupervised learning
  • deep learning and Transformer-based models
  • natural language processing (NLP)
  • generative AI and signal-based prediction

Open-Source & Tool Development

• Developed VolumePeeler, a FIJI plugin for 3D tissue visualization and quantification used in biomedical research
• Built pipelines for rock lithology classification, microscopy image analysis, and biomedical imaging
• Promoted reproducibility and adoption of AI workflows across academic and applied settings
• Developed tools supporting interdisciplinary collaboration between engineers, clinicians, and researchers

Technical Skills

• Programming: Python, R, MATLAB
• Libraries & Tools: PyTorch, TensorFlow, scikit-learn, OpenCV, FIJI/ImageJ
• Practices: model evaluation, reproducibility, ethical AI, data privacy

Collaboration & Impact

• Worked with multidisciplinary teams including engineers, clinicians, and domain experts
• Translated technical results into actionable insights for decision-making
• Experience delivering solutions under real-world constraints, including data limitations and stakeholder requirements
• Strong focus on deployable, real-world systems, not only research prototypes
• Proven ability to bridge academic research and applied industry solutions