Looking for a job

mohamed mostafa

@mohamedmostafa12

Senior Backend Engineer at Incorta who cut memory usage by 33% and improved data-platform performance.

At Incorta, I led the architecture and delivery of a migration from off-heap to on-heap storage for a data platform running on terabytes of RAM. I optimized G1GC and implemented custom SoftReferences, cutting memory usage by 33% and improving filtering, distinct operations, and search execution.

At Valeo, I owned the design and delivery of a Google Cloud-based NoSQL Java application supporting timesheets for 37k employees and 10k+ projects. Partitioning its main table by month reduced monthly reporting costs by about 88% and increased performance by 3x on BigQuery storage.

Earlier at Valeo, I developed a multi-target tracking system, designed a custom DBSCAN variant, and built and optimized a production simulation platform using ROS and Gazebo. At the Egyptian Army, I built a national asset monitoring and management system and a point-of-sale system.

Experience

Work history, roles, and key accomplishments

Current

Senior Backend Engineer

Current

Incorta

Feb 2022 - Present (4 years 8 months)

Led architecture and delivery of a 9-month migration from off-heap to on-heap storage for a data platform, optimizing G1GC and implementing custom SoftReferences to cut memory usage by 33% and improve filtering and distinct operations by 60% and search execution by 25%. Designed a robust fix for a critical production issue and led requirement gathering, design, implementation, and code review for

Java G1GC SoftReferences Memory Optimization Performance Optimization System Architecture Code Review Requirements Gathering

Principal Software Engineer

Valeo

Jan 2020 - Jan 2022 (2 years)

Owned requirements, design, implementation, test planning, and code review for a Google-Cloud-based NoSQL Java application supporting timesheets for 37k employees and 10k+ projects. Mentored 2 senior and 3 junior engineers, controlled cloud costs, and partitioned main table by month reducing query costs by ~88% and increasing performance by 3x on BigQuery storage.

Java Google Cloud Platform NoSQL BigQuery Cloud Cost Optimization Mentorship Test Planning Code Review

Senior Algorithm Engineer

Valeo

Jul 2017 - Jan 2020 (2 years 6 months)

Designed and implemented a custom version of DBSCAN for multi-target tracking, developed a multi-target tracking system, and built a production simulation platform using ROS and Gazebo. Optimized a boundary fusion and lane tracking algorithm to reduce memory by 75% and enhance performance by ~20%.

DBSCAN Multi Object Tracking ROS Gazebo Algorithm Optimization C++Python