Engineered scenarios, which are templated SQL queries for use by non-technical users. Scenarios are used to flag suspicious activity within our standardized data models.

Developed pre-computation data pipeline jobs to optimize the run-time of scenario executions

Feature engineer for a domain specific language built on top of SQL for detecting fraudulent activity for businesses.

Scaled back-end data pipelines to process millions of claims data.

Python, Spark, Airflow, AWS, Docker, Looker/Presto/Hive (Distributed SQL w/ metastore)

Built two stateless python AI systems that were used within an AWS micro-service stack as a scalable customer service chatbot

Instant-AI

induces conversational rule-sets from historical conversations through semi-supervised intent matching.

Provides baseline of chatbot rule-sets.

AI-Engine: Context Free Grammar chatbot

Front-end dashboard allows agents to add and configure existing rule-sets which govern the chatbot

AI-Engine uses these rulesets to engage in conversations

Brought two data pipelines into production

Distributed RabbitMQ reader with JSON SQL Loader ETL process

Kafka, Samza, ElasticSearch for real-time stream processing of live engagement metrics

Winner of internal hackathon: Implemented Recommendation Engine prototype, which was then brought into production.

Consulted as a data engineer on a team of four for a Fortune 500 company and provides modular REST data services with fault-tolerant distributed caching using Redis, Zookeeper, HSQLDB, and Hibernate

(acquired by Apple)