Data & AI Engineer | Cloud Data Platforms • Applied ML/NLP • RAG

I build reliable cloud data pipelines, analytics-ready datasets, machine learning systems, and source-grounded AI applications using Python, SQL, Azure, AWS, Databricks, PySpark, dbt, PostgreSQL/pgvector, scikit-learn, PyTorch, and Transformers.

I’m Rihua Van Steenburgh, a Data and AI Engineer focused on building reliable cloud data platforms, analytics-ready datasets, applied machine learning systems, and source-grounded RAG applications.

I am pursuing a Master of Science in Information Technology with a Data Analytics concentration at Middle Georgia State University, expected in December 2026. I also hold an Associate of Applied Science in Website Design/Development, which helps me connect data and AI backends with practical user-facing applications.

My portfolio combines cloud data engineering, analytics engineering, NLP model evaluation, and Generative AI application development. I focus on reproducible workflows, honest evaluation, testing, documentation, and clear technical communication.

• Cloud Data Engineering: API ingestion, Azure Data Factory, ADLS Gen2, Databricks, PySpark, Delta Lake, AWS S3, ECS/Fargate, Redshift Serverless, and dbt

• Applied Machine Learning & NLP: scikit-learn, TF-IDF, Linear SVM, PyTorch, Transformers, DistilBERT, leakage-safe evaluation, and error analysis

• Generative AI & RAG: document ingestion, chunking, embeddings, PostgreSQL/pgvector, vector retrieval, source-grounded answers, citations, and safe no-answer behavior

• Analytics Engineering & Reliability: SQL, dimensional modeling, analytics marts, data validation, pytest, GitHub Actions, Docker, evaluation, runbooks, and technical documentation

• Languages: Python, SQL, R, JavaScript, TypeScript

• Cloud & Data Platforms: Azure Data Factory, ADLS Gen2, Databricks, AWS S3, Redshift Serverless, ECS/Fargate, EventBridge, CloudWatch

• Processing & Modeling: PySpark, Delta Lake, dbt, PostgreSQL, dimensional modeling, star schema

• AI / RAG: document ingestion, chunking, embeddings, PostgreSQL/pgvector, vector search, retrieval-augmented generation, cited answers

• Orchestration & CI/CD: Airflow, GitHub Actions, workflow automation, validation checks

• BI & Tools: Power BI, DAX, Docker, Git/GitHub, Jupyter Notebook, VS Code

• Web & APIs: REST APIs, JSON, CSV, WordPress

Hybrid RAG application that ingests curated NYC 311 documentation, stores embeddings in PostgreSQL/pgvector, retrieves cited context, routes approved analytics questions, and presents grounded answers through a Streamlit interface.

Tech: Python, PostgreSQL, pgvector, embeddings, vector search, RAG, Streamlit, Docker, pytest, GitHub Actions

Leakage-safe eight-class CFPB complaint-routing study comparing a TF-IDF and Linear SVM benchmark with a frozen DistilBERT challenger. The project includes group-aware splits, model evaluation, selective routing, and Human Review policies.

Tech: Python, scikit-learn, TF-IDF, Linear SVM, PyTorch, Transformers, DistilBERT, model evaluation

Azure lakehouse pipeline using Azure Data Factory, ADLS Gen2, Databricks, PySpark, SQL, and Delta Lake to produce validated Bronze, Silver, Gold, fact, dimension, and analytics-mart outputs.

AWS batch data pipeline using Docker, ECS/Fargate, EventBridge, S3, Redshift Serverless, SQL, and dbt with data-quality tests, CI checks, runbooks, and cloud execution proof.

I am upgrading CivicLens from a local Hybrid RAG prototype into a more complete AI application with:

• a versioned FastAPI backend;
• an optional real LLM provider;
• citation validation and safe provider-error handling;
• repeatable RAG and LLM evaluation reports;
• query, retrieval, and feedback logging;
• Dockerized Streamlit, API, and PostgreSQL/pgvector services;
• optional bounded agent and analytics-tool routing.

• 💼 LinkedIn
• 🌐 Portfolio
• 🧑‍💻 GitHub

Thanks for stopping by! ✨