Dexter Taylor
Principal Data Engineer at Concerto HealthAI- Claim this Profile
Click to upgrade to our gold package
for the full feature experience.
Topline Score
Bio
Experience
-
ConcertAI
-
United States
-
Biotechnology Research
-
700 & Above Employee
-
Principal Data Engineer
-
Jan 2019 - Present
New York, NY Designing and building Pythonic data pipelines to feed Concerto's oncology analytics and modeling tools. Creating automation stacks in Ansible and Terraform for auto-deploying pipelines and provisioning and tuning data management services.
-
-
-
Binary Machines, LLC
-
United States
-
IT Services and IT Consulting
-
Data Engineer
-
Nov 2017 - Present
New York Built a web-based warehouse management application and custom data pipeline for a leading NYC artwork rental shop to synchronize between the physical warehouse, commerce site, and accounting platform. Used Mercury (Python data pipeline framework), Eavesdroppr (Postgres async transaction monitoring framework), PostgreSQL, and the Snap microservice framework to synchronize internal inventory records with third-party e-commerce and accounting. Used SiCX (Pythonic HTML screen scraping toolkit based… Show more Built a web-based warehouse management application and custom data pipeline for a leading NYC artwork rental shop to synchronize between the physical warehouse, commerce site, and accounting platform. Used Mercury (Python data pipeline framework), Eavesdroppr (Postgres async transaction monitoring framework), PostgreSQL, and the Snap microservice framework to synchronize internal inventory records with third-party e-commerce and accounting. Used SiCX (Pythonic HTML screen scraping toolkit based on BeautifulSoup) and Mercury to scrape and ingest inventory records from legacy web catalog into the new system. Show less
-
-
-
-
Senior Software Engineer
-
May 2018 - Oct 2018
New York Creating web services on AWS and Google Cloud to support AMC's Shudder and Sundance Now content channels. Prototyping data pipelines using Python, Kinesis, MongoDB, and Elasticsearch to support in-house analytics requirements. Designed and built an API wrapper around AWS Cognito (using the Snap web framework) to provide user-auth functionality for AMC's next-gen content delivery system.
-
-
-
Riparian LLC
-
United States
-
Pharmaceutical Manufacturing
-
1 - 100 Employee
-
Lead Platform Engineer
-
Mar 2017 - Nov 2017
United States Responsible for designing and building data pipelines and services for reporting and data analytics used in the pharmaceutical industry. Under construction: PRAXIS, a pipeline architecture using Kafka as an enterprise data bus between data ingest, ETL services, and analytics clusters. Using Snap (Python microframework) for REST API construction, and Mercury (formerly METL) for ETL and data transfer between Kafka, PostgreSQL, and Couchbase datastores. Cloud-based architecture using AWS… Show more Responsible for designing and building data pipelines and services for reporting and data analytics used in the pharmaceutical industry. Under construction: PRAXIS, a pipeline architecture using Kafka as an enterprise data bus between data ingest, ETL services, and analytics clusters. Using Snap (Python microframework) for REST API construction, and Mercury (formerly METL) for ETL and data transfer between Kafka, PostgreSQL, and Couchbase datastores. Cloud-based architecture using AWS with Ansible and CloudFormation for provisioning and orchestration. Show less
-
-
-
Tictail (acquired by Shopify)
-
Sweden
-
Technology, Information and Internet
-
1 - 100 Employee
-
Senior Data Engineer
-
Oct 2016 - Feb 2017
New York Responsible for designing and implementing ETL data pipelines between core business systems and analytics. Designed and built TDX (Tictail Data eXchange) pipeline for populating OLAP star schemas using METL (pipeline construction toolkit) and Snap Python frameworks. OLAP tables in AWS Redshift, Couchbase and redis on EC2 for intermediate record storage, Chef for server configuration, Ansible for pipeline "last-mile" self-provisioning. Built journaling framework for data pipeline… Show more Responsible for designing and implementing ETL data pipelines between core business systems and analytics. Designed and built TDX (Tictail Data eXchange) pipeline for populating OLAP star schemas using METL (pipeline construction toolkit) and Snap Python frameworks. OLAP tables in AWS Redshift, Couchbase and redis on EC2 for intermediate record storage, Chef for server configuration, Ansible for pipeline "last-mile" self-provisioning. Built journaling framework for data pipeline self-reporting and failure recovery; built REST endpoint using Snap for pipeline monitoring. Designed and tested experimental SHaPE (Segment Hash Parallel Execution) architecture for improved pipeline performance. Show less
-
-
-
-
Senior Data Engineer
-
2014 - Sep 2016
NYC Data handling architecture for Vulcan energy efficiency metrics project. Designed and built Vulcan virtual data center on AWS, using Ansible for provisioning. VDC supports web presence using HAProxy and Nginx. Internal & external DNS via Route53. Designed and built Pulse message-handling system, using Kafka for pub-sub queuing and Cassandra for storage. Created relational data model for simulation and modeling of heating and cooling systems. Created Soma cluster maintenance utility… Show more Data handling architecture for Vulcan energy efficiency metrics project. Designed and built Vulcan virtual data center on AWS, using Ansible for provisioning. VDC supports web presence using HAProxy and Nginx. Internal & external DNS via Route53. Designed and built Pulse message-handling system, using Kafka for pub-sub queuing and Cassandra for storage. Created relational data model for simulation and modeling of heating and cooling systems. Created Soma cluster maintenance utility using Zookeeper, Python and Boto. Spun up microservices using Snap (Python framework based on Flask). Show less
-
-
-
-
Senior Software Engineer
-
2013 - 2014
NYC Wrote, tested, and deployed Python scripts for data center migration for boinc.com music streaming service. The task was to migrate ~300TB of content data from one data center to another while the system was under load. Designed and wrote a family of scripts called Cyclops to watch a source filesystem for changes and log them to a repository; other scripts in Cyclops would consume the change messages and generate rsync commands to move the data. Used MongoDB as the producer-consumer queue.
-
-
-
Marigold
-
United States
-
Marketing Services
-
500 - 600 Employee
-
Team Lead
-
2011 - 2012
NYC HERCULES crash project for analytics overhaul. Assembled a team and architected and built a prototype system to replace the NoSQL-based analytics system at the Sailthru ESP. Hercules is a hybrid SQL/ NoSQL system with an OLAP star schema at its core, populated by Python and Java event listeners consuming business messages from a Redis queue, with Serpentine handling the front end. In initial testing the Hercules system outperformed the existing MongoDB-based system by two orders of magnitude.
-
-
-
-
Senior Consultant
-
2010 - 2011
NYC Clean-sheet redesign and build of Bifrost video workflow management software. Bifrost is a database-driven, browser-based app which allows Healthination’s digital content manager to organize, tag, and insert ads into Healthination’s original video content for encoding and packaging by their Avalon encoding platform. Written using Serpentine and MySQL on the back end, HTML/CSS/ Javascript on the front end (data visualization sections done using Protovis). Server-side “directory-watch” shell… Show more Clean-sheet redesign and build of Bifrost video workflow management software. Bifrost is a database-driven, browser-based app which allows Healthination’s digital content manager to organize, tag, and insert ads into Healthination’s original video content for encoding and packaging by their Avalon encoding platform. Written using Serpentine and MySQL on the back end, HTML/CSS/ Javascript on the front end (data visualization sections done using Protovis). Server-side “directory-watch” shell scripts also written in Python. Show less
-
-
-
-
Director
-
2008 - 2010
Brooklyn, NYC Taught one-on-one production workshops. Designed and built motion picture lighting equipment in custom-equipped machine shop: Mercury (low-cost fluorescent), Apollo (compact high-intensity dimmable LED), Argus (remote-dimmable 2x3 halogen bank light).
-
-
-
Innovest
-
Research Services
-
1 - 100 Employee
-
Consultant
-
2006 - 2008
Designed and wrote system for import and processing of financial modeling information for Innovest’s green-investment-profiling system. MySQL DB for data storage; object mapping and metadata handling code written in Java.
-
-
-
New York Film Academy
-
United States
-
Higher Education
-
700 & Above Employee
-
Instructor
-
2004 - Dec 2006
Production sound/postproduction sound instructor. Responsible for generating lesson plans, assembling materials, teaching lessons, and accompanying students on field workshops.
-
-
-
-
Senior Software Architect
-
2003 - 2004
Senior Software Architect. Responsible for software design driving the cutomer application SmartServeMan- ager (SSM). SSM is the route by which Poindexter customers input ad campaign information into the Poindexter Optimization Engine, an automated mechanism for optimizing online ad campaigns. • Refactored original codebase. • Designed and implemented database-level API for revised security system to allow greater flexibility. • Captured requirements from in-house and offsite users… Show more Senior Software Architect. Responsible for software design driving the cutomer application SmartServeMan- ager (SSM). SSM is the route by which Poindexter customers input ad campaign information into the Poindexter Optimization Engine, an automated mechanism for optimizing online ad campaigns. • Refactored original codebase. • Designed and implemented database-level API for revised security system to allow greater flexibility. • Captured requirements from in-house and offsite users to drive product enhancements. Show less
-
-
-
-
Chief Technology Officer
-
2001 - 2003
Chief Technology Officer. Responsible for all software and hardware acquisition, development, integration, and deployment.
-
-
-
SapientRazorfish
-
United States
-
Business Consulting and Services
-
500 - 600 Employee
-
Senior Technologist
-
1998 - 2000
Responsible for software engineering tasks related to existing web sites and those under development for Razorfish clients.
-
-
-
-
Senior Developer
-
1997 - 1998
Responsible for systems design and implementation on a variety of projects relating to GlobeComm's core service offering, which was electronic mail. All coding was done in C++ on Solaris, using Oracle 7.3.x for data storage.
-
-
-
Bloomberg
-
United States
-
Financial Services
-
700 & Above Employee
-
Software Developer
-
1995 - 1997
Software Developer. Responsible for various projects from inception through requirements capture, prototyping, and development.
-
-