Avatar
😀

Organizations

README.md

Iván is a biomedical research software engineer in the Biomedical Applications Group at the Pittsburgh Supercomputing Center (PSC). He contributes to pioneering scientific initiatives such as the Human BioMolecular Atlas Program (HuBMAP), the Brain Image Library (BIL), and SenNet, and he is part of the engineering team for Bridges-2, PSC’s flagship advanced research computing platform. His work centers on advancing computational methods, reproducibility, and scalable workflows in biomedical research.

He pursued graduate studies at Carnegie Mellon University (CMU) with funding from the National Institutes of Health. Before joining PSC, he worked in the Murphy Lab within CMU’s Computational Biology Department and collaborated with departments including Biomedical Engineering and Biological Sciences.

Beyond his PSC role, Iván serves as a scientific computing consultant for the Ray and Stephanie Lane Center for Computational Biology in Carnegie Mellon University’s School of Computer Science, and also served as an undergraduate program manager for the Mellon College of Science, where he acted as a bridge between students and administration to support underrepresented groups in the sciences.

Iván was the longtime host of Barrio Latino, Pittsburgh’s first Spanish-only radio show, which aired for over 25 years. He remains deeply committed to the Hispanic community through his leadership in the Pittsburgh Spanish Language Group and his role (at times) organizing Hispanic Heritage Month events with institutions like CMU and the University of Pittsburgh.

An open-source enthusiast, Iván is passionate about repurposing older technology, optimizing computational workflows in biology, and developing innovative solutions to complex research challenges. Outside of work, he enjoys cooking and learning to play the ukulele. His community contributions — including relief efforts following Hurricane María — have been recognized in CMU publications, underscoring his dedication to advocacy, innovation, and cultural engagement.

3d-modeling access acnh adventures aging ai ai-agents algorithms anatomy of the ear android animation apptainer argentina art astrology atlas automation baby yoda backend barrio latino bash bbq bedtools best practices big-data bil bioformats bioinformatics blockbench blogging bon-jovi books brain-image-library brain-initiative breakfast breast-cancer bridges-2 bridges2 brookline teen outreach center bst cbd10 cell-geometry cellorganizer certification cfde clas claude cli client-server clp clustering cmu coding-exercises cody collaboration comic commercials community computational-biology computer-science computing containerization containers cooking cooklang copa america covid-19 crafts creative creative-commons croissant cursor data-cleaning data-curation data-ecosystem data-engineering data-management data-portal data-repository data-structures datasets dbscan dimensionality-reduction disk-usage distributed-systems dj the cut docker doctor who documentary education em-algorithm experiments fair family fermentation festival filesystems finland food food-experiments fortnite fortran gaming gaussian-mixture-model gear generative generative-models git github github-actions github-copilot gmm golang gratitude hierarchical hobby hpc hubmap huggingface hugo humor hurricane maria image-analysis imagemagick images instagram iris k-means knitting korean lab lab meeting lacu lamp latex latino latinx lclaa life linear-regression linux llm love lustre machine-learning macos macosx matlab matplotlib media-server meetup metadata mexican microscopy minecraft mlcommons mmw multi-omics museum music naruto national-weather-service navidad networking neuroscience new-year nextflow nih nintendo nlp nsf ols ome-tiff omero omero.importer omero.insight onboarding ontology open-access open-data open-meteo open-science open-source openai organ-donation pandas parquet pascualina pca personal pet-projects pets phd-thesis pittsburgh pixar plain-text plex plugins podcast preprint presentation productivity programming psc publishing puerto rico python r r-project radio rag random rebirth recipes reference-atlas reflection regression remote-work reproducibility research resources rust sacnas salsa science scikit-learn scripting senescent-cells sennet single-cell singularity skimr slurm smithsonian social software spanish spatial-atlas spatial-simulation standards statistics storage super bowl 2020 tacos taskwarrior tensorflow terminal text-to-speech tinycards tools training travis-ci trees tts tutorial twitter university of pittsburgh utilities valentines-day vegan vegetarian version-control video video games vocabulary vscode walter mercado weather webservers workflows wrct

Popular posts

  1. What Hugging Face is and what it is for, and how it relates to Croissant metadata files and TensorFlow in a modern ML workflow.

    machine-learning huggingface datasets tensorflow croissant

  2. A reference glossary of the key components that make up a Lustre parallel file system—MDS, MDT, OSS, OST, MGS, clients, LNet, and the terms you need to read a Lustre layout.

    storage hpc filesystems lustre

  3. A definition of Lustre, how its architecture achieves massive parallel bandwidth, and how it compares to other file systems common in HPC—GPFS/Spectrum Scale, BeeGFS, NFS, and Ceph.

    storage hpc filesystems lustre

  4. A formal definition of a file system, an intuitive TL;DR, and a look at parallel file systems with a focus on Lustre.

    storage linux hpc filesystems

  5. What a Croissant file is, why it matters for machine-learning datasets, and a complete worked example describing a real, published HuBMAP CODEX spleen dataset.

    datasets metadata croissant hubmap machine-learning

  6. A formal definition of tokens and tokenization, and why language models process text as tokens rather than words or characters.

    ai machine learning nlp llm

Post activity