Agent lifecycle orchestration for Apache Airflow
Cross-cloud integrations that turn an Airflow DAG into a control plane for managed AI agent runtimes.
I build data platforms and agentic systems that turn complex ideas into useful, reliable products.

Platforms I build with
Selected work
Cross-cloud integrations that turn an Airflow DAG into a control plane for managed AI agent runtimes.
A working troubleshooting loop for AI agents on AWS Bedrock AgentCore, Google Vertex AI and Microsoft Azure AI Foundry.
Strengthened Magpie's agent-assisted PR review so dependency findings require complete constraint evidence and remediation follows each repository's policy.
Added merged operators for modifying DMS replication tasks and reloading selected target tables, including deferrable waits and end-to-end AWS validation.
Speaking · Official Apache Airflow Monthly Town Hall
Presented at Apache Airflow’s official monthly gathering for project leaders, contributors and practitioners.
Explore the talk and referencesWriting
Open source · AWS
Agents · Airflow
Architecture · AWS
Data platform · AI
RAG · Bedrock
About

I want what I learn to become a path others can walk to get further, faster — just as I once stood on the shoulders of giants.
My background is in database systems at Universidad Tecnológica Nacional. Today I design Data & AI platforms at Mendel and contribute production capabilities across Apache Software Foundation projects.
In Apache Airflow, I authored merged integrations for managed AI agent platforms across AWS, Google Cloud and Microsoft Azure, plus operator work across core AWS services including DMS, ECR and Glue.
In Apache Magpie, I strengthened agent-assisted code review with evidence-based dependency findings and repository-specific remediation.
Also recognized at the 2025 Harvard Health Systems Innovation Lab Hackathon for building high-value health systems through AI.
Read the full profileCredentials
Connect