I design and build production-grade AI systems that operate across cloud, distributed compute, edge devices, and real-world infrastructure. My focus is on AI-native architectures, forecasting systems, accelerated machine learning, and intelligent automation across networked and physical systems.

Currently pursuing an M.S. in Applied Artificial Intelligence at the University of San Diego (2027 anticipated).

• M.S. in Applied Artificial Intelligence (University of San Diego, 2027* anticipated)

I am an AI/ML & Solutions Architect building intelligent, AI-native systems at the intersection of cloud infrastructure, edge computing, and real-world networks. My work spans AI-RAN, Private 5G, GPU-accelerated analytics, and Physical AI — from pipeline design to production deployment and beyond.

🧑‍💻 AI-RAN 5G KPI Forecasting

• Goal: Build a GPU-accelerated ML forecasting system for 5G RAN KPI prediction with full MLflow experiment tracking.
• Tech Stack: Python, PyTorch, XGBoost, RAPIDS/cuDF, MLflow, Distributed Compute
• Outcome: Delivered a reproducible, production-grade forecasting pipeline with accelerated training and experiment traceability across distributed compute environments.
• GitHub Repository

🧑‍💻 Private 5G RAN Pipeline

• Goal: Design an end-to-end telemetry pipeline for Private 5G RAN data ingestion, transformation, and storage.
• Tech Stack: Python, Apache Spark, Airflow, dbt, Parquet, AWS/Azure
• Outcome: Scalable ingest → transform → parquet architecture suitable for RAN analytics at production volume, enabling downstream ML and reporting workflows.
• GitHub Repository

🧑‍💻 Telecom Churn EDA & ML

• Goal: Build an explainable churn prediction system for telecom customers using interpretable ML techniques.
• Tech Stack: Python, scikit-learn, XGBoost, SHAP, Pandas, Jupyter Notebook
• Outcome: Delivered a churn classification model with SHAP-driven feature attribution reporting, supporting Responsible AI practices and stakeholder transparency.
• GitHub Repository

🧑‍💻 QPSK Wireless Link Simulator

• Goal: Simulate and analyze the performance of QPSK digital wireless communication links.
• Tech Stack: Python, NumPy, Sionna, Matplotlib
• Outcome: Produced BER performance analysis across varying SNR conditions, demonstrating communication system behavior and establishing a foundation for ML-integrated RAN experimentation.
• GitHub Repository

🧑‍💻 Autonomous Parallel Parker System

• Goal: Design and build a physical AI system capable of performing autonomous parallel parking using IR sensors.
• Tech Stack: Python, Embedded C, IR Sensors, Raspberry Pi / Jetson, Edge AI
• Outcome: Delivered a working sensor-driven decision loop enabling real-time autonomous parking maneuvers, demonstrating embedded inference at the edge.
• GitHub Repository

🧑‍💻 Breast Cancer Detection — Agentic ML Workflow

• Goal: Build an explainable cancer detection classifier with an agentic reporting layer that delivers SHAP-based insights via email.
• Tech Stack: Python, scikit-learn, SHAP, LangChain, Jupyter Notebook
• Outcome: Model achieved strong classification accuracy with automated agentic email delivery of SHAP explanations, demonstrating interpretable AI with actionable reporting pipelines.
• GitHub Repository

📫 Reach me via email or on LinkedIn if that's your preference.