02 / PROJECTS
Built to learn.
Designed to work.
Selected work in infrastructure engineering and applied machine learning.
Showing 2 projects
FEB – MAY 2026 / DISTRIBUTED SYSTEMS
Real-Time Infrastructure Monitoring & Incident Response
Turning infrastructure event streams into actionable incident insights.
Built a Python and Kafka monitoring platform processing over 10,000 events per minute across 50+ simulated services.
- Traced cascading failures through service dependencies and root-cause analysis.
- Stored incident and service information in PostgreSQL.
- Deployed an AWS-based response engine to prioritize severity and recommend remediation, with detection under five seconds.
PythonKafkaPostgreSQLAWS
JAN – MAY 2025 / MACHINE LEARNING
Stroke Prediction Pipeline
A recall-focused approach to identifying high-risk cases.
Created an end-to-end Python and scikit-learn pipeline covering preprocessing, feature engineering, training, and evaluation.
- Optimized Random Forest classifiers to achieve 85% recall on high-risk cases.
- Addressed class imbalance and worked to reduce false negatives.
- Built reproducible evaluations to compare experiments and guide feature and model selection.
Pythonscikit-learnRandom ForestModel evaluation