Data Science · AI Engineering · Graph Neural Networks · Cloud Analytics
Senior Data Solutions Architect · Spatio-Temporal AI Specialist · AWS Certified
Enterprise engineering discipline applied to the frontier of AI and machine learning. I architect resilient, end-to-end data solutions — from cloud pipelines to Graph Neural Networks — translating complexity into clarity and systems into strategic insight.
Who I Am
My foundational career was defined by the "how" of software. At companies like TCS — building for global clients like Lloyds Bank — I focused on the critical infrastructure that keeps businesses running: systems support, rigorous testing, and front-end engineering with React.js. This era taught me that the most advanced technology is only as good as its stability. I learned to navigate complex enterprise environments and ensure that code was genuinely resilient and user-centric.
While my early career was spent ensuring systems worked, I found myself drawn to why they worked — and to modeling the most complex system of all: the human mind. This curiosity drove my transition into Data Science. My long-term vision is to bridge high-level AI engineering and neuroscience, building the foundation today that will allow me to contribute to neural pattern recognition and brain-computer interfacing tomorrow.
I am seeking roles that value professional maturity combined with a relentless drive to solve human-scale problems — where engineering discipline meets the ambition to push the boundaries of what AI can achieve.
Multi-objective optimization and spatial equity modeling for dynamic MTA transit pricing — addressing annual revenue loss through ST-GNN and Pareto-optimal fare calculation.
ST-GNNs using PyTorch and DGL to model network conversations across time and detect advanced C2 threats that traditional static models cannot capture.
Evaluating LLM architectures via LMSYS Arena; monitoring NeurIPS and ICML to integrate cutting-edge advances into production workflows.
Delta Lake, Apache Iceberg, and Airflow DAGs — building governed, incremental, production-grade platforms for the next generation of analytics.
Career
Delivered end-to-end analytics platform solutions, from raw data ingestion to governed datasets and operational reporting, using Python and automated pipelines across the full SDLC.
Engineered a Dockerized Python fraud detection engine, processing referral entries with custom fraud logic and UTC normalization across global timezones to ensure temporal data integrity.
Built HubSpot Deal Closure probability models using Logistic Regression and XGBoost, evaluated via ROC-AUC to drive measurable improvements in sales forecasting accuracy.
Performed Sentiment Analysis and Flight Risk forecasting using Random Forest models to surface organizational communication trends and proactively identify risks.
Designed and maintained scalable React.js component libraries powering high-traffic enterprise applications, achieving measurable reductions in load times through performance-first architecture.
Architected and delivered full-stack feature modules across complex multi-tier applications, collaborating with UX designers and backend engineers to build accessible, production interfaces.
Managed the complete promotion and release lifecycle across dev/test/prod environments — enforcing CI/CD best practices and automated testing pipelines for zero-downtime deployments.
Championed code quality and engineering standards through rigorous peer reviews, documentation, and knowledge transfer, improving team velocity.
Designed and configured AWS cloud infrastructure (EC2, S3, RDS, VPC, IAM, Route53, CloudWatch, Auto Scaling) to support enterprise analytics workloads.
Implemented Snowflake as the enterprise data warehouse, optimizing compute usage via multi-cluster warehouses; developed ETL pipelines using Python and Airflow DAGs.
Designed semantic data models and DAX measures in Power BI and Tableau to standardize business metrics and enable governed self-service analytics.
Built Scala-based Spark applications for large-scale data cleansing, enrichment, and aggregation — processing hundreds of millions of records.
Automated financial services via the Loan Module Portal for Lloyds Bank, streamlining database workflows and reducing manual intervention for critical global banking infrastructure.
Developed and optimized SQL queries and stored procedures to improve reporting performance and system reliability.
Expertise
Key Projects
Architected and deployed FareFlex, an end-to-end solution addressing annual revenue loss. The system replaces static flat-fares with an Intelligent Elastic Fare model that dynamically adjusts prices based on real-time infrastructure health and passenger density.
ETL pipeline ingesting OMNY rows, GTFS-Realtime feeds, and MTA Needs Assessment. Spatial joins via GeoPandas to establish Social Equity pricing floors.
ST-GNN modeling the station network as a directed graph. SciPy multi-objective solver for Pareto-optimal fare per trip.
Dynamic Pricing API and high-fidelity Mobile UI/UX. Real-time Micro-Fare incentives during major service disruptions.
Tableau Command Center for stakeholders to simulate scenarios — signal failures, weather surges, event peaks.
Developed a Spatio-Temporal Graph Neural Network using PyTorch Geometric to identify the behavioral fingerprint of network threats and C2 traffic patterns.
Architected Airflow DAGs to migrate ADLS CSV datasets to Delta Lake/Iceberg with incremental loading strategies. Integrated Great Expectations for automated quality audits.
Benchmarked Multimodal AI Agents using LMSYS Arena and monitored NeurIPS and ICML to integrate state-of-the-art architectures into research workflows.
Built a machine learning pipeline using Random Forest to validate NASA's Kepler stellar objects and classify exoplanet archetypes.
Engineered a Dockerized Python engine to process referral entries with custom fraud logic, flagging invalid rewards with real-time monitoring.
Credentials
Pace University, Seidenberg School of CSIS
Focus: Robotics and Control Systems
Amazon Web Services
Amazon Web Services
Graph Algorithms & GDS Library
Internet of Things Certification
EdX · Certified
Let's Connect
Open to data science, AI/ML engineering, and research-driven roles — especially those at the intersection of applied AI and human-scale problems.