Gabriel Sumabat
Senior Data Engineer, Data Science and Machine Learning @Numerator
Signup · Get unlimited contacts
WORK HISTORY
Senior Data Engineer, Data Science and Machine Learning @Numerator
Toronto, ON, CA
Data Engineer building both Pipelines and Platforms focused on DS/ML Products. Infra: Terraform, Kubernetes, Helm, AWS, ECS, Kinesis, Docker, GithubPipelines/Data Warehouse: Python, SQL, Airflow, Pyspark/EMR, Snowflake, Postgres, KinesisPlatform: Python, FastAPI, JS, Svelte, Fast-AgentArchitecture: Event Driven, Microservice, Serverless• Championed and led the development of a data platform for capturing advanced financial reports and e-commerce metrics to be used as weights for training market panel models resulting in proper data governance, improved model stability and increased the number of raking benchmarks. • Scaled Data Science proof of concepts into production pipelines responsible for producing market share valuations at a daily granularity for an average of 3.2B combinations on a weekly basis using Snowflake and Airflow. • Reduced a mission critical Spark pipeline overall cost by 55% by reducing on demand node cost from $3.06/h to $1.63/h and runtime from 3h22 to 2h45 by rewriting key joins and optimizing cluster memory utilization.• Managed resilient infrastructure using Terraform and Helm to build Kubernetes (EKS) clusters and ECS and reduced monthly infrastructure cost by 17% by optimizing instance selection, AWS retention and centralizing resource usage.• Implemented a custom Role Based Access Control using JWTs to improve system stability, governance and traceability improving UI performance and reducing system load for Apache Airflow.• Built a Real Time Streaming Pipeline ingesting 50M events a day to drive advanced user analytics for core business operations providing immediate feedback for changes implemented in the Mobile App.• Built a prototype for an LLM powered pipeline for extracting specific retailer calculations from companies financial statements utilizing OpenAIs models with custom MCP servers to integrate with internal systems.
EDUCATION
George Brown Polytechnic Continuing Education
Certificate for Allen-Bradley PLC, Electrical Engineering Technologies/Technicians
Ryerson University
Bachelor's Degree, Electrical Engineering
SKILLS
ABOUT GABRIEL SUMABAT
Senior Data Engineer specializing in building scalable data platforms and machine…
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.