A reusable data foundation for risk, products, analytics and AI, not one-off pipelines. Ingestion, medallion warehouse layers, domain marts and operational distribution from a single architectural spine.

AWSRedshiftGlueS3KinesisdbtTerraform

Large-scale company risk-data architecture that ingests, enriches, normalizes and distributes company-level intelligence across countries. Built as a reusable risk pool for underwriting, products, analytics and intelligent systems, and today the data foundation beneath Cowbell's OMNI AI agents.

Risk DataEnrichmentMulti-CountryData Marts

Cowbell · 2024 - Present

GlueFlux

Read case study

Metadata-driven processing framework. Pipelines defined in YAML so teams onboard Spark, Python and dbt workloads without rebuilding orchestration, deployment and operational patterns every time.

YAMLGlueECSSparkdbtPlatform

Redshift as a reusable enterprise warehouse: Bronze, Silver, Gold and domain marts with dbt models, separating ingestion, normalization, business modeling and consumption.

RedshiftdbtMedallionData Marts

Moving curated data to regional applications, search and APIs while respecting residency, PII, latency and operational independence. Distribution as a platform capability, not a copy job.

Multi-RegionDMSPostgreSQLGovernance

Trusted data and platform foundations that enable intelligent product experiences: governed flows, grounding and operational access for agents and AI-driven insurance workflows.

Agentic AIRAGGroundingLLMs

Healthcare research, via Nineleaps · 2021 - 2022

Research Study Exports

Read case study

A Kubernetes job per study. Cloud SQL replica for the read, Python workers for the research tables, dbt on BigQuery for the heavy pass, then a checked package in the recipient bucket.

PythonKubernetesCloud SQLBigQuerydbt

API Gateway and a document-store stream land in S3 through Kinesis and Firehose. A rules API and a field library write the curated prefix. S3 Batch replays history. Redshift is what analytics query.

KinesisFirehoseS3KMSRedshift

Ext: Uber, via Nineleaps · 2018 - 2019

External Data Platform

Read case study

Built and coordinated the ETL pipelines that brought external data into Uber: ZoomInfo, Zendesk, Uber Eats and weather on H3, shaped for the app and for the models, including the foundation under price prediction. Spark and Kafka, at terabyte scale.

SparkKafkaPythonExternal DataData Quality

Data and model-computation pipelines behind Blue Yonder's supply-chain AI: preparing enterprise data, running ML models for demand forecasting and delivering predictions at scale on Azure.

AzurePySparkSynapseMachine LearningCI/CD

ATMS and UDL: Python and AWS frameworks that use compiler-style models to generate every feature combination a processor must pass before release, so coverage is designed, not guessed.

PythonAWSTest AutomationCompiler DesignCombinatorics

American Megatrends · 2014 - 2017

Master Data & Release Platform

Read case study

Built the central master data system, with regional replication that respected PII, amisvnservice in C++ for internal communication, and the subscription notices and email around a BIOS release.

C++MDMReplicationNotificationsRelease automation