Skip to content
View Michael-Gurule's full-sized avatar
0 followers · 3 following
  • Denver, Colorado, USA

Block or report Michael-Gurule

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
michael-gurule/README.md

Hi, I'm Michael Gurule

Data Scientist | Former Strategy Consultant | Production ML Engineer

Transitioning from 8 years of Fortune 500 consulting to data science roles, with a focus on production-grade ML systems and business-driven analytics.

About Me

  • Currently building production ML systems and advanced analytics projects
  • MS Data Science + AWS Certified Data Engineer + Google Advanced Data Analytics
  • Former Senior Strategy Consultant at Sedgwick and JS Held
  • Based in Colorado
  • Seeking Senior Data Scientist / ML Engineer roles

What sets me apart: I combine deep technical ML skills with 8 years of Fortune 500 client experience. I don't just build models—I build production systems that deliver measurable business impact.

Currently under active development

SENTINEL Multi-Intel Platform

A production-grade machine learning platform demonstrating advanced sensor fusion techniques for Defense applications. Integrates Overhead Persistent Infrared (OPIR) thermal detection with Radio Frequency (RF) geolocation algorithms for real-time threat detection and tracking.

Deployed Features

Multi-Sensor Fusion Engine

  • Implements Kalman filtering with covariance-weighted measurement fusion
  • Mahalanobis distance data association for multi-target tracking
  • Real-time track quality assessment and uncertainty quantification

RF Geolocation Algorithms

  • TDOA/FDOA positioning using nonlinear least-squares optimization
  • Hybrid time-frequency solver for improved accuracy
  • GDOP/PDOP computation for sensor geometry assessment
  • Mixed-altitude sensor network design (500m-2000m) for optimal 3D positioning

Deep Learning Pipeline

  • 1D CNN for thermal event classification (5 classes)
  • PyTorch training pipeline with early stopping and learning rate scheduling
  • 10,000+ physics-based synthetic training samples

Features Under Development

  • Interactive Streamlit dashboard for real-time track visualization
  • 3D sensor network and track position plotting
  • Multi-Hypothesis Tracking (MHT) for improved ambiguity resolution
  • Extended Kalman Filter (EKF) for nonlinear motion models
  • Additional Sensor Modalities EO/IR (Electro-Optical/Infrared) SAR (Synthetic Aperture Radar)

Portfolio Projects

Natural Disaster Risk Intelligence Platform

Real-time multi-hazard risk assessment integrating USGS, NASA, and NOAA data sources.

  • Tech Stack: Python, FastAPI, Streamlit, Pandas, Plotly
  • Highlights: Multi-source API integration, geographic risk scoring, production error handling
  • Business Value: Portfolio risk assessment for real estate and insurance industries

Real-Time Fraud Detection System

Production ML pipeline with cost-optimized decision thresholds delivering $474K monthly savings.

  • Tech Stack: Python, XGBoost, FastAPI, Streamlit
  • Highlights: Custom cost functions, sub-100ms API latency, interactive monitoring
  • Business Value: Optimized precision-recall tradeoff based on real business costs

Marketing Attribution with Causal Inference

Advanced causal methods (PSM, DiD, Uplift) to measure true incremental marketing impact.

  • Tech Stack: Python, scikit-learn, statsmodels, DoWhy, CausalML
  • Highlights: Propensity score matching, difference-in-differences, CATE estimation
  • Business Value: Separates correlation from causation in marketing ROI analysis

Technical Skills

Languages & Core:
Python SQL R

ML & Data Science:
scikit-learn XGBoost Pandas NumPy

Production & APIs:
FastAPI Streamlit Docker

Cloud & Data:
AWS PostgreSQL

Specializations:

  • Production ML Systems & MLOps
  • Causal Inference & Experimental Design
  • Business Analytics & Strategy
  • API Development & Data Engineering

Certifications

  • MS Data Science - In Progress
  • AWS Certified Data Engineer – Associate
  • Google Advanced Data Analytics Professional Certificate

Let's Connect

LinkedIn Email Medium

Open to opportunities in data science, ML engineering, and analytics roles. Particularly interested in companies solving real-world problems with data.

From michael-gurule

Pinned Loading

  1. fraud-detection-system fraud-detection-system Public

    A production-ready machine learning system for detecting credit card fraud in real-time with sub-100ms latency. Built with XGBoost, FastAPI, and Streamlit.

    Python 1

  2. marketing-attribution-causal-inference marketing-attribution-causal-inference Public

    Causal inference system for marketing attribution using PSM, DiD, and uplift modeling with interactive Streamlit dashboard

    Python 1

  3. disaster-risk-platform disaster-risk-platform Public

    A production-grade data pipeline that ingests real-time data from government APIs (USGS, NASA, NOAA) to assess natural disaster risk across earthquakes, wildfires, and severe weather.

    Python 1

  4. sentinel-multi-intel-platform sentinel-multi-intel-platform Public

    Python 1