Ryan Dunker

Selected Projects

Real-time Sentiment Analysis API

Production-ready sentiment analysis API handling 1000+ requests per minute with sub-50ms latency. Built with FastAPI and deployed on AWS with automated monitoring.

Python FastAPI Docker AWS

ML Model Monitoring Dashboard

Comprehensive monitoring system for tracking model performance, data drift, and system health in production environments with real-time alerting.

Python Streamlit PostgreSQL Prometheus

Computer Vision Pipeline

End-to-end object detection pipeline with model training, ONNX optimization, and cloud deployment. Achieved 40% improvement in inference speed.

PyTorch ONNX TensorRT GCP

LLM Fine-tuning Framework

Custom framework for fine-tuning large language models with efficient training, evaluation pipelines, and deployment automation.

PyTorch Transformers LoRA MLflow

A little about myself

Hi, I’m Ryan Dunker. I help HP make confident decisions by transforming raw data into dependable systems. Combining data science with data engineering, I create pipelines, tools, and workflows that solve real-world problems efficiently.

Strongest in:

Python PyTorch TensorFlow FastAPI Docker AWS

Familiar with:

Kubernetes GCP React SQL MLflow Terraform

Tools:

Git VS Code Jupyter Postman GitHub Actions