Skip to content

Experience

  • Mibanco — Credicorp

    Jun 2026–present

    Artificial Intelligence Intern · Lima, Peru

    DatabricksPythonSQLAzure AI

    • Built a parallelized audio-ingestion and speech-to-text pipeline on Databricks (Python, SQL) using Azure AI model endpoints, for integration into production workflows at Peru’s largest microfinance bank
    • Designed LLM-based interpretation of transcribed speech; orchestrated multiple Azure AI models to extract meaning from recorded audio
    • Coordinated the AI team’s project planning while automating routine engineering tasks with LLM-assisted tooling

  • TRUE (Trustworthy and Responsible) AI Lab — Oregon State University

    Nov 2025–Jun 2026

    Undergraduate Research Assistant — LLM Fault Resilience · Corvallis, OR

    PyTorchHugging FaceSlurm

    • Evaluated robustness of 1B–8B Llama 3 models to hardware bit-flip faults: built a gradient-guided attack framework and contributed evaluation code to the lab’s fault-injection harness (PyTorch, Hugging Face), run at scale on an A100 Slurm HPC cluster
    • Ran the initial investigation and early results for a draft-model-gated decoding defense and early-warning detection metrics against targeted bit-flip attacks
    • Benchmarked 7 fault-tolerance defense methods against 2 classes of adversarial bit-flip attacks (stealth and damage) on GSM8K
    • Co-authored manuscript on cross-scale model resilience, a speculative-decoding-based fault detection and correction defense; contributed mitigation implementations and trade-off analysis

  • Oregon State University Foundation

    Jan 2026–present

    Digital Technology Student Assistant · Corvallis, OR

    Data visualizationSitefinity CMS

    • Built data visualizations and summary reports from operational datasets for institutional reporting; maintained production web pages in Sitefinity CMS

  • Universidad de San Martín de Porres

    Jul 2024–Aug 2024

    Python Course Assistant · Lima, Peru

    Python

    • Mentored students on Python assignments and debugging, graded with detailed rubrics, and led 5+ study sessions