Rohit Malik

Designed and implemented LLM-based investigative search systems operating over 500M+ records Built document intelligence pipelines for structured entity extraction from legal documents Implemented semantic search and vector-based retrieval for identity matching Designed SCD Type 2 data models ensuring historical correctness and auditability Delivered production AI systems with role-based access control and compliance constraints

Designed scalable data pipelines using Databricks, PySpark, and SQL to support analytics and downstream AI workloads Orchestrated workflows using Apache Airflow, improving data delivery timelines Built curated, high-quality datasets used for search and ML experimentation Improved data accuracy and consistency through validation and reconciliation Supported data platforms used by international clients

Built robust data processing pipelines using AWS Glue, PySpark, and Databricks Integrated multiple heterogeneous source systems into centralised datasets Designed structured data models to support analytics and early AI initiatives Developed reusable ETL frameworks and documentation to improve maintainability Mentored junior engineers on data engineering best practices