17+ years in data. Five industries, banking to healthcare.
Senior Data Engineer | AI Engineer. Five years building cloud data platforms on AWS and Azure: ETL/ELT on Glue, Step Functions, Airflow and Databricks, lakehouse on Apache Iceberg, LLM-based agents for pipeline observability. Before that, twelve years of analytics and data science in banking and payments.
Nine companies, in reverse order. Every entry maps to a LinkedIn role and a stack you can dig into on a call.
Data Engineer & AI Engineer. Contracting through Napoli Data LLC: distributed data pipelines and cloud data platforms across AWS and Azure, ETL/ELT orchestration with Glue, Step Functions, Airflow and Databricks, and LLM-based assistants for pipeline observability with LangChain, Anthropic Claude, Amazon Bedrock and Ollama for local inference.
Senior Data Engineer. Designed and operated a multi-tenant serverless data standardization platform on AWS. Carrier files land into raw, cleansed and standardized layers on Apache Iceberg, orchestrated by Step Functions, Glue and Lambda. Onboarded new carriers end to end, from file contract to validated production load. Incremental loads with SCD Type 2 and Iceberg MERGE, automated post-load validation, and production incident response. Wrote a local LLM analyzer (LangChain, Ollama, DeepSeek) that ranked root-cause hypotheses for on-call.
Data Engineer. Distributed pipelines on AWS (Glue, Lambda, Step Functions, Athena) and Airflow on Kubernetes with KubernetesPodOperator. Real-time ingestion from REST APIs and Kafka; automated ETL from third-party APIs into Azure SQL; hybrid flows across Azure Data Factory, Synapse and Databricks. Tuned Redshift queries and PostgreSQL RDS for scalability and cost. Automated monitoring with LLM agents built on LangChain and Claude.
AWS Data Engineer. Data pipelines on Lambda, Glue and RDS Aurora, with Airflow, Step Functions and EC2 for large volumes; Pandas and PySpark for analysis at scale; S3 and Athena for storage and querying; CloudWatch for monitoring. Re-engineered an ETL process running every 15 minutes on EC2, RDS and S3.
AWS BI Data Engineer. Designed, implemented and maintained the cloud data architecture on AWS: EC2, S3, Lambda to automate ETL and maintenance tasks, Glue crawlers to catalog data in S3, and Redshift and Athena for analytics, with query tuning and schema-design practices for speed and accuracy.
Senior Data Engineer. Large-scale ETL pipelines on AWS (Python, Lambda, Glue, Athena, S3) turning raw MySQL and S3 data into features for data scientists and reports for business analysts. Put machine-learning solutions into production (environment setup, Bitbucket version control, unit testing in Flask/Python) and improved monitoring with CloudWatch and EventBridge.
Data Scientist Project Leader. Big Data & Analytics at a multi-brand payment processor: led the data-science team's projects, planned enterprise-wide data solutions, built and monitored predictive models, and defined business metrics and delivery timelines.
Data Analyst (Credit Risk) → Senior Data Scientist (Marketing & BI). Propensity models for insurance and loans, a cross-sell recommendation engine, RFM and unsupervised customer segmentation, a real-time recommender on Oracle and an NLP ReMarketing chatbot. In Credit Risk: a lending qualification engine and automated qualification of the credit-card portfolio. Measurable sales lift, material call-center savings.
Case study overview →Data Analyst. Credit Risk Management (2013-2016): credit scoring and evaluation models for individuals and companies. Built, updated and optimized scores and mass ratings. Commercial Operations Support (2009-2013): statistical studies on credit-card claims and report automation.
Where the data comes from, what gets built, what ships.
Airflow and Step Functions execution logs, Glue job failures, health-monitor output, cloud alarms.
LLM agents with LangChain and Claude (or Ollama for local inference when data can't leave) that read logs and monitor output, summarize pipeline state and rank root-cause hypotheses.
Prioritized incidents for on-call with a root-cause hint and next action. Fewer surprises, better pipeline visibility.
Postgres, S3 dumps, SaaS APIs, Kafka streams. Mixed schemas, mixed cadences, no shared dictionary.
ETL/ELT on AWS or Azure, Spark on Glue/Databricks for heavy load, Iceberg layers with SCD Type 2 and MERGE for incremental history. Cost and query performance tuned from day one.
Curated tables in Redshift, Athena or Snowflake. BI-ready and validated after every load: row counts, duplicates, orphans, column shift.
Transactional warehouse, CRM attributes, behavioral signals, campaign history.
Feature engineering plus classical models (logistic, gradient boosting, RFM, unsupervised segmentation) or NLP when text dominates. Put into production and monitored.
Scored customers piped to CRM, campaigns, credit decisions and churn watchlists. Decisions move from gut to evidence.
Not a laundry list. Tools used in production, grouped by where they sit in the data path. Highlighted = primary in the last two years (Proactiviti, CLARA Analytics).
+ 4 more in databricks-examples · dbt-examples (ML/MLflow, cost optimization, testing, performance)
A 30-minute call to see if there's a fit. No pitch.
Defined scope and timeline. Typically 4-12 weeks.
Senior part-time capacity, embedded in your team. Typically 2-3 days a week.
Architecture, code review, technical decisions. Monthly retainer, async plus a weekly call.
Buenos Aires (UTC−3) · 6+ hours overlap with US Eastern · Contracting through Napoli Data LLC · I reply within one business day
Third-party proof: read the recommendations on LinkedIn →