Cloud Architect at Wright-Patt Credit Union — designing and operating cloud infrastructure in a regulated financial environment. Building toward ML Engineering and AI Engineering roles, with production projects spanning LLM safety, federated learning, differential privacy, real-time MLOps, clinical NLP, and computer vision.
| Certification | Issuer |
|---|---|
| AWS Certified Machine Learning – Specialty | Amazon Web Services |
| AWS Certified Cloud Practitioner | Amazon Web Services |
| GCP Professional Data Engineer | Google Cloud |
| Snowflake SnowPro Advanced: Data Engineer | Snowflake |
| dbt Analytics Engineering Certification | dbt Labs |
| Domain | Technologies |
|---|---|
| ML / Deep Learning | PyTorch · XGBoost · scikit-learn · BERT · BioBERT · HuggingFace · Grad-CAM · Federated Learning (Flower) |
| LLMs & AI Agents | LangChain · LangGraph · RAG · DP-SGD · Opacus · Prompt Engineering · OWASP LLM Top 10 |
| MLOps | MLflow · SageMaker Pipelines · Airflow · Evidently AI · Drift Detection · Model Registry · A/B Testing |
| Data Engineering | Kafka · Apache Spark · dbt · Redis · Airflow · SQL lineage · ETL/ELT |
| Cloud & Infra | AWS (SageMaker · Redshift · S3 · CloudWatch · DynamoDB) · GCP · Docker · FastAPI |
| Privacy & Compliance | Differential Privacy · Federated Learning · HIPAA · GDPR · ECOA / Fair Lending · Presidio PII Redaction |
| Repo | Description | Stack |
|---|---|---|
llm-safety-auditor |
LLM red-teaming: 250+ adversarial attacks, OWASP LLM Top 10 scoring, FastAPI audit reports | sentence-transformers · FastAPI |
finance-llmops-platform |
RAG over SEC 10-Ks + earnings calls: citation-grounded answers, hallucination detection | LangChain · MLflow · Streamlit |
agentic-pipeline-healer |
LangGraph agent monitors Airflow DAGs, diagnoses failures, auto-applies fixes | LangGraph · Airflow · FastAPI |
clinical-note-llmops |
HIPAA clinical NLP: Presidio PII scrubbing → BioBERT NER → LLMOps | Presidio · spaCy · FastAPI |
prompt-ops |
Git-style LLM prompt versioning & A/B testing — MLflow for prompts | MLflow · FastAPI |
evidently-llm-sentinel |
Production LLM monitoring: semantic drift, hallucination risk, Grafana dashboards | Evidently · Grafana |
| Repo | Description | Stack |
|---|---|---|
mlops-retraining-pipeline |
Drift-triggered retraining: KS-test gates → Welch t-test → A/B validation → promotion | Airflow · MLflow · SageMaker |
kafka-stream-feature-store |
Real-time ML feature store: Kafka → Redis, sub-60s feature freshness | Kafka · Redis · FastAPI |
ml-drift-monitoring |
Production ML monitoring: feature drift, prediction drift, data-quality alerts | scikit-learn · Evidently |
chexpert-pathology-classifier |
Chest X-ray multi-label: DenseNet121+EfficientNet-B4 ensemble, 92% AUC, Grad-CAM | PyTorch · SageMaker |
federated-credit-risk |
3-institution federated credit risk via Flower FedAvg — zero raw data sharing, ECOA | Flower · PyTorch |
icu-mortality-predictor |
30-day ICU mortality from first 24h MIMIC-III — XGBoost + SHAP, AUC >0.85 | XGBoost · SHAP · FastAPI |
| Repo | Description | Stack |
|---|---|---|
high-traffic-ticket-engine |
Redis distributed-lock ticket engine — 10K concurrent users, 0 oversells | Redis · FastAPI |
distributed-etl-ecommerce |
PySpark medallion ETL + dbt star schema + Airflow + Great Expectations | PySpark · dbt · Airflow |
data-lakehouse-delta |
Delta Lake lakehouse: bronze/silver/gold medallion with ACID transactions | Delta Lake · Spark |
redshift-wlm-optimizer |
ML predicts Redshift query cost, auto-routes WLM queue — 85% runtime reduction | XGBoost · AWS Redshift |
sql-to-dag-compiler |
Oracle SQL/PLSQL + dbt → Airflow DAGs with lineage export (Mermaid/DOT/JSON) | Python · Airflow · dbt |