Data Platform Architect & AI Systems specializing in modern cloud data platforms, enterprise data governance, and production-grade agentic AI systems.
I focus on engineering high-throughput AWS data architectures, building scalable self-service infrastructure, robust lineage tracking, and automated decision engines across complex transactional and unstructured domains.
| Domain | Technologies & Capabilities |
|---|---|
| Languages & Core | Python, SQL, FastAPI, REST APIs, JSON Schemas |
| Data Infrastructure & Cloud | AWS (S3, Athena, Glue, QuickSight), Databricks (Delta Lake), Apache Airflow, dbt, MySQL, Terraform (IaC), Neo4j |
| AI Systems & Orchestration | Agentic AI (CrewAI, LangChain), RAG Evaluation, Vector Databases, XGBoost, SHAP, AWS Bedrock / PartyRock |
| Data Governance & Quality | OpenMetadata, OpenDataDiscovery (ODD), Data Contracts, Automated Lineage, Data Quality Frameworks, Collibra |
- Data-Platform — Multi-layer enterprise data platform featuring dbt models, AWS S3 / Athena query layers, Terraform IaC, and Apache Airflow orchestration.
- FinPulse-Market_agent — Multi-agent financial intelligence system orchestrating autonomous research and market analytics workflows.
- Ecommerce-domain-nlp — Scalable text classification and semantic parsing framework for multi-category e-commerce product catalogs.
- payment-fraud-ml — End-to-end real-time anomaly detection pipeline using XGBoost, SHAP explainability, and FastAPI microservices.
- nlp-payments-lab — Enterprise NLP framework covering classic TF-IDF, BERT fine-tuning, and LLM-driven retrieval agents.
- Network-graph-intelligence — Graph database modeling using Cypher for complex relational network intelligence.
- OpenMetadata: Authored comprehensive guides and architectural patterns for ML model governance, training data lineage, custom metrics tracking, and automated alert workflows (PR #31058).
- ODD Platform: Developed end-to-end data lineage tracking integration scenarios for open-source data discovery tools.