Hi, I'm Marcelo Villarreal
Where data science meets software development, and innovation is born. Welcome to my portfolio.

I'm a Staff Engineer specializing in full-stack product development and cloud-based data systems. I build production web applications, customer-facing workflows, backend services, and the data pipelines behind them.
View ResumeMy Experience
I've worked in various companies, leveraging my expertise to drive innovation and deliver results. My experience spans multiple technical domains, primarily in developing and optimizing python pipelines (ML models, data analytics tools, etc.), and designing and implementing full stack web applications.ClimateAi
Feb 2026 - PresentStaff Engineer
Leading development of ClimateAi's weather intelligence platform, spanning customer-facing workflows and AWS pipelines for data ingestion and model training.
ClimateAi
Mar 2025 - Feb 2026Data Scientist L3
Drove 0-1 full-stack development of customer-facing features, from interface design through backend integration and deployment.
ClimateAi
Feb 2024 - Mar 2025Data Scientist L2
Developed statistical and machine learning models for crop-yield forecasting and productionized them for use in climate-risk applications.
ClimateAi
Jun 2023 - Feb 2024Junior Data Scientist
Developed physics-based crop-yield forecasting models and integrated them into agricultural planning tools.
Dropin
Jul 2022 - Jun 2023Full-Stack Developer
Built and maintained full-stack components, including an API for automated delivery-guide generation and an AWS Lambda route optimizer built on Google's Distance Matrix API to reduce costs and transit times.
Monofile
An AI-powered markdown editor and cloud filesystem built for humans and agents alike, with a full agent-facing API and MCP server.
Yeyar
A real estate platform that allows users to browse live projects and interact with developers to find out about the latest investment opportunities.
Full-Stack Web & Mobile
From interface to database — Next.js and Expo apps backed by typed APIs and Postgres schemas, built as monorepos with shared packages so one feature ships to web, mobile, and desktop from a single codebase.
Data & ML Engineering
Terabyte-scale weather and geospatial pipelines — source ingest through Zarr aggregation, debiasing, and QA into a Spark and Iceberg lakehouse, orchestrated with Airflow and AWS Batch — plus the crop-yield forecasting and downscaling models that run on top of them.
Agentic Engineering
Building for agents as first-class users — MCP servers and permission-scoped tool APIs, the context and skills that make a codebase legible to them, and multi-agent workflows that fan out triage, review, and documentation in parallel.
Contact Me
Have a question or want to work together? Send me a message using the form.
Email: contact@cloudportfol.io
