RAG Works Better When Your Data Is Already Tidy
Every retrieval project I've watched go badly went badly for the same reason.
Read morePrincipal Data Engineer · Azure and applied AI
Jeremias Jaymez
I'm from Córdoba, Argentina, living in The Netherlands. Working at Rockdata, focused on data platforms, platform reliability, and applied AI.
From the blog
Recent notes on data engineering, platforms and AI.
Dagster Pipelines Need Clear Ownership
A job failed at four in the morning. Whose was it?
Read moreCourses
What I've learned building data platforms, turned into something you can follow step by step.
Lakehouse Masterclass
Build a complete open-source lakehouse on your laptop: Iceberg tables on S3-compatible storage, a Nessie catalog with Git-style branches, Spark, Dagster, and a local RAG over your own docs. Twelve labs, no cloud account needed.
The labs are already open on GitHub. I'm building a full course on top of them.
Challenges
Two projects with hard constraints that taught me more than the easy ones did.
Secure platform
Built a secure workstation for data scientists and analysts at a financial institution. Designed networking, storage, compute, data ingestion, availability, and software maintenance to meet compliance and performance demands. Helped align efforts with architects, users, and the Product Owner.
Accurate data delivery
Worked for a consultancy company as a contractor for a USA non-profit supporting first-time moms. Delivered business rules, dashboards, and a web app to validate bronze, silver, and gold layers. Supported HIPAA compliance and improved consistency for manually entered data.
About
Data, ops and development. That mix is usually why I can see where a problem actually starts.
Data Engineer
Data Engineer with experience in Azure and Databricks. Applied Machine Learning during my Master's thesis and personal projects. Started as BI Developer/Analyst building DWH, ETL, and dashboards.
SQL, Python, PySpark, Databricks, Synapse, ADF, MicroStrategy, Pentaho, Power BI, modeling.
DevOps Engineer
Infrastructure as Code, pipeline reliability, and DataOps. Databricks administration, securing components, workflows, and policy configuration.
Azure Pipelines, Bicep, Bash, PowerShell, Terraform, CI/CD. Networking (vNets, VPN, Private Endpoints).
Software Engineer
Arsysco products and client systems, built end to end: Domium, a property management platform, and an internal app for a law firm in Córdoba that calculates court-ordered interest. Started with freelance full stack work in .NET and C#.
Python, FastAPI, SQLAlchemy, React, TypeScript, Azure Container Apps, Azure SQL. Earlier: .NET Core and C#.
Education
Information Systems Engineer with a Master's in Information Systems Engineering (2023) and a Master's in Artificial Intelligence (2025).
Founder of Arsysco, the studio behind Domium.
Software for businesses since 2014, from Córdoba. Domium, our property management platform, is live.
Full timeline on LinkedIn Data engineering examples on GitHub
Contact
Open to interesting work and collaborations. Any of these will reach me.