AI & Data Engineer @ Bank Alfalah Ltd. | Trustworthy AI in Finance
Building robust, scalable, and secure data ecosystems is the foundational mandate of modern enterprise finance.
Processing millions of financial transactions at scale is a solved engineering problem; ensuring the absolute integrity of those algorithmic decisions is not.
My engineering trajectory is dedicated to advancing Trustworthy AI -architecting the core science that dictates explainable, equitable, and secure financial technology.
- AI-Based Risk Automation: Engineered Hugging Face NLP models and XGBoost/PyTorch behavioral models for Suspicious Transaction Reports (STR), prioritizing mathematical explainability and significantly reducing false positives.
- Petabyte-Scale Infrastructure: Orchestrated the enterprise deployment of Cloudera Data Platform (CDP 7.x) across a 12-node, 105TB production cluster, maintaining 99.9% architectural uptime via distributed Linux workload balancing.
- Data Governance & Zero Trust: Architected strict Apache Ranger security policies for Hive and HBase environments, enforcing enterprise-wide RBAC and achieving zero unauthorized exposures across 105TB of distributed storage.
- Compliance & Scale: Designed scalable PySpark and REST API ingestion pipelines increasing data load efficiency by 40%, while engineering RPA checkpoints to enforce 2025 TBML regulations.
Big Data Ecosystem: Apache Spark | Kafka | Hadoop | Hive | HBase | Impala | HDFS | YARN | CDP 7.x
AI & Machine Learning: PyTorch | Hugging Face | XGBoost | MLflow | NLP | RPA
Languages & Frameworks: Python | Scala | PySpark | SQL | Spark SQL | Bash | FastAPI
Infrastructure & Datastores: Linux (RHEL) | Apache Ranger | PostgreSQL (pgvector) | MS SQL Server | Oracle DB | MySQL
