I build systems and lead teams that make massive data move fast and run efficiently.
Currently, I'm a Consultant and General Manager at the National Center for AI (SDAIA), directing engineering teams on petabyte-scale data pipelines and data quality. Before that, I spent years in the database trenches at Couchbase (as Principal Software Engineer) and Apache AsterixDB, designing hyper-efficient storage engines and query optimization primitives. I thrive at the intersection of deep-tech systems engineering, scalable architecture, and high-performance team leadership.
Returned to the consumer side to lead and supervise a talented engineering team handling petabyte-scale data projects. Responsible for driving the center's data quality initiatives and optimizing data processing and wrangling pipelines that serve a wide range of analysts and scientists across different domains.
Worked on optimizations for the query and storage engines of the Big Data analytical platform. Building advanced scanning, pruning, and transactional storage primitives.
Contributed to analytical query and storage optimizations for document stores, improving scan throughput and data layout.
Added native query capabilities for Parquet files from external cloud storage engines (e.g. AWS S3) in Couchbase Analytics. Developed projection and filter pushdown logic to minimize network roundtrips.
Actively contributing to open-source database internals, with focus on LSM storage engines, columnar document representation, page-level compression, and compiler optimizations.
Worked at the Center for Complex Engineering Systems (CCES) – a joint collaboration between KACST and the Institute for Data, Systems and Society (IDSS) at MIT. Designed big data analytics frameworks, including the open-source AsterixDB-Spark Connector, CityDynamics, and Integrated Transportation Systems.
Developed firmware and collection software connecting electric and water smart-meters to data collection networks for R&D projects (ADDAD4, Water Smart Meter).
Thesis: Towards Analytics-optimized Document Stores
Supervisor: Michael J. Carey
Thesis: NUMA-aware Multicore Matrix Multiplication
Supervisor: Isaac D. Scherson
Graduated with a GPA of 4.55/5.0, earning Second Class Honors.
Proposed a new mechanism to leverage Log-Structured Merge (LSM) lifecycle events to automatically infer schema and semantically compact self-describing semi-structured records. Implemented in Apache AsterixDB along with page-level compression, achieving a 9.8x reduction in storage size and query speedups of the same factor.
Extended the Dremel columnar format to support document stores' flexible, nested data model. Designed and implemented two layouts for organizing and storing schemaless data in LSM-based storage systems, optimizing query scans over nested, sparse document fields.
Explored runtime query compilation techniques for document stores where value types are unknown until query time. Leveraged Oracle Truffle to compile execution sub-trees in a Java-based document store, showing major speedups over standard vectorized batch processing models.
Recognized as a high performer and one of the top achievers at Couchbase Inc. for outstanding engineering contributions.
Awarded full graduate scholarship from the King Abdulaziz City for Science and Technology (KACST) for Ph.D. studies at UC Irvine.
Awarded full graduate scholarship from King Abdullah Scholarship Program for M.Sc. studies at UC Irvine.
Awarded Second Class Honor for outstanding academic performance and high GPA at King Saud University.