Juan Carlos De Miguel Rodríguez
Summary
Senior Data Scientist with 5 years of experience in developing and implementing machine learning systems for industrial condition monitoring and predictive maintenance. Designed a scalable ML training framework that achieved over 90% reduction in runtime and managed end-to-end deployments using AWS SageMaker. Expertise in Python, SQL, Snowflake, and MLOps, with practical experience in deploying LLM applications and AI agents.
Overview
2
2
Languages
5
5
years of professional experience
Work History
Data Scientist
Caterpillar
Chicago, Illinois
09.2021 - Current
- Delivered Cat Digital's first production-scaled analytics product (Automated Fault Codes) as sole developer, building the end-to-end training and inference pipeline on Snowflake, AWS SageMaker Processing, and Apache Airflow; drove $250M sales in 2024, $300M in 2025, and set the blueprint reused by subsequent Cat Digital ML products.
- Architected and led development of Hyperscaling, Cat Digital's scaled ML training/deployment framework for condition-monitoring models, authored the foundational data-processing layer in Python (Polars, Pydantic, Pytest) and served as primary pull-request reviewer for the platform, now used by 30+ Data Scientists across the org and powering products responsible for $100M+ in annual sales.
- Led end-to-end production deployment of all Sense Repair models, Fuel System, Aftertreatment two-step, Turbo, and their expansions, managing DEV, QA, and PROD validation and co-building the inference pipeline on AWS SageMaker, S3, Airflow, and Snowflake; contributed $80M+ in Closed Won revenue in 2025 and built a dedicated deployment pipeline to serve models on CAT Foresight.
- Identified and eliminated compute bottlenecks in the Hyperscaling codebase, refactoring data-processing logic in Polars and introducing caching layers, reducing end-to-end pipeline runtime by >90% and materially lowering AWS compute cost.
- Prototyped internal AI agents on AWS Bedrock, delivered as VS Code extensions, to automate repetitive ML-development tasks (feature engineering, data exploration, model optimization), saving Data Scientists weeks of manual effort per project.
- Defined and coordinated the Hyperscaling 2.0 roadmap in Azure DevOps, authoring user stories, running Agile/Scrum ceremonies, and aligning multiple developers to ship scoped improvements including lead-quality features and inference-pipeline enhancements.
- Conducted extensive data exploration and cleaning through SQL, Python, and Snowflake. Contributed to data quality validation during Helios migration, an enterprise-wide effort.
- Defined and coordinated the Hyperscaling 2.0 roadmap in Azure DevOps, authoring user stories, running Agile/Scrum ceremonies, and aligning multiple developers to ship scoped improvements including lead-quality features and inference-pipeline enhancements.
- Mentored 5+ interns and new hires (including SHPE-program mentees), recognized as the team's go-to resource for technical guidance and troubleshooting; represented Cat Digital as an SHPE recruiter for three consecutive years, supporting the org's hiring and diversity pipeline.
Software Engineer / Data Science Intern
Caterpillar
Chicago, Illinois
06.2021 - 08.2021
- Prototyped LSTM-based time-series models in Python (TensorFlow/Keras) to classify component health from high-frequency sensor telemetry, improving condition-monitoring accuracy over rule-based baselines by 20%.
- Built severity-score and Remaining-Useful-Life (RUL) algorithms in Python for predictive maintenance of critical machine components (Engines, Aftertreatment systems), feeding downstream dealer-facing dashboards used to prioritize service actions.
- Executed data-cleansing and feature-engineering pipelines (SQL, Pandas) on multi-source machine telemetry, unlocking previously unusable historical data for downstream model training and analytics.
Education
Master of Science - Computer Science
Illinois Institute of Technology
Chicago, IL08-2021
Master of Science - Industrial Engineering
Universidad Politécnica De Madrid
Madrid, Spain08-2021
Bachelor of Science - Industrial Engineering
Universidad Politécnica De Madrid
Madrid, Spain11-2019
Skills
- Machine Learning: Supervised learning, gradient boosting (XGBoost, LightGBM), Random Forest, feature engineering, hyperparameter tuning, multi-step and ensemble models, time-series modeling, model explainability
- MLOps & Production: End-to-end model deployment, inference pipeline design, AWS SageMaker, Apache Airflow, MLflow (experiment tracking, model registry), production model monitoring, drift detection
- Data Platforms & Cloud: Snowflake, AWS (SageMaker, S3, Lambda), ETL/ELT pipeline design, large-scale SQL
- Programming: Python (Pandas, Polars, NumPy, Pydantic, pytest, scikit-learn), SQL, Git, GitHub, Docker, CI/CD (GitHub Actions)
- Deep Learning & Generative AI: Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), prompt engineering, LSTM networks, LangChain, MCP servers, Keras
- Data Visualization: Plotly, Matplotlib, Seaborn, Streamlit, Jupyter, Power BI, Tableau, website presentations
- Leadership & Process: Azure DevOps, Agile / Scrum (sprint planning, story authoring, backlog grooming), cross-functional stakeholder communication, roadmap definition, mentorship, code review leadership
Leadership & Community
- SHPE (Society of Hispanic Professional Engineers) active member; represented Cat Digital at national/regional conventions to recruit engineering and data science talent.
- Data4Good, Turning Pointe. Volunteered through Caterpillar's Data4Good initiative, partnering with a non-profit supporting individuals with autism to build analytics tooling and data pipelines that improved their reporting and decision-making.
Languages
English
Full Professional
Spanish
Native/ Bilingual
Timeline
Data Scientist
Caterpillar
09.2021 - Current
Software Engineer / Data Science Intern
Caterpillar
06.2021 - 08.2021
Master of Science - Computer Science
Illinois Institute of Technology
Master of Science - Industrial Engineering
Universidad Politécnica De Madrid
Bachelor of Science - Industrial Engineering
Universidad Politécnica De Madrid