Skip to content

Hello, I'm Hope Tatenda Mutema

Data · Actuarial · AI

A passionate data analyst, aspiring actuary, and curious AI enthusiast.

I love turning messy datasets into clear narratives, designing risk-aware models, and experimenting with agentic AI systems that help organizations make smarter, faster decisions.

Recently, I have been exploring deep research agents, financial and insurance analytics, and ways to bring production-ready AI into real products and workflows.

Welcome to my corner of the internet where analytics, risk, and creative AI experiments meet.

GitHub LinkedIn
AI Enthusiast

Major Projects

Civil Society & CRPD: NLP Text Analysis

Synthesized 337 reports from OHCHR. Used a custom NLP pipeline to map regional policy variations and vulnerability patterns across global disability reporting.

NLP Python / R
View Code

Insurance Lapse: Predictive Modeling

Analyzed 228k records using Random Forest and LASSO. Optimized detection of at-risk policyholders by 2x through threshold tuning for revenue protection.

Churn Analytics LASSO / RF
View Analysis

Drivers of Mortality: Tableau BI Study

OLS regression across 3,183 counties identifying income inequality and poverty as primary death predictors over healthcare access.

BI / Health Tableau
View Repo

Other Projects

Corruption & HDI Visualizer

Global correlation analysis between Corruption (CPI) and Human Development (HDI). Built a publication-ready scatter plot using R to highlight governance trends and outliers.

Economist-style aesthetic theme
Log-linear trend regression
R Language ggplot2 data.table ggthemes
Deep Research Agent
with Pydantic AI

Built a multi-step research orchestrator using Pydantic AI and GPT-5-mini. Generates structured, evidence-based reports on stock tickers and complex topics.

Intent-aware stock/general search
Parallel dives with cited evidence
Python GPT-5-mini Structured Output
Health Insurance
Risk Modeling

Built cost models in Python to quantify how age, BMI, smoking status, and region influence medical charges and actuarial risk tiers.

Smoker vs non-smoker risk gaps
Age & BMI impact on charges
Python scikit-learn Ridge Regression
Netflix Movie Ratings
Regression Analysis

Applied multiple regression in R to analyze how genre, viewing hours, release timing, and seasonality impact global ratings.

Multi-Variable Content Analysis
OLS Predictive Modeling
R Language ggplot2 Stats Models

Pathfinder AI

Sustainable Impact Challenge

Built an AI-powered platform using Agentic workflows with n8n to automate data extraction and deliver personalized academic recommendations for DC students.

Dual-Enrollment CTE Optimization
AI for Educational Equity
n8n Workflows Agentic AI Automation
Ratings & Global
Reach Analysis

Applied Chi-square tests & regression in R to explore statistical relationships between movie ratings and worldwide availability patterns.

Categorical Statistical Testing
Global Distribution Patterning
R Language Chi-Square Data Analysis
Netflix Availability
& Ratings Patterns

Applied simple linear regression in R to analyze how global availability impacts movie ratings across diverse international content.

Global Performance Trends
Distribution-to-Rating Correlation
R Language Regression Patterns
Spotify Popularity
Feature Analytics

Analyzed 1M+ Spotify tracks in R to quantify how audio features and temporal trends drive song popularity using regression models.

Multi-Feature Popularity Drivers
Large-Scale Dataset Engineering
R Language Correlation Data Viz
Latest Updates

Recent News & Insights

Exploring the frontiers of data science, actuarial innovation, and AI.

AI
APRIL 2026

Building Production-Ready Agentic AI Systems

Deep research agents automating financial analysis workflows at scale.

Read full article →
Data Science
MARCH 2026

Mixed-Methods NLP Analysis of UN Reports

TF-IDF, NER, and lexicon-assisted qualitative coding of 500+ CRPD reports.

Read full article →
Actuarial
FEBRUARY 2026

Risk-Aware Modeling & AI Integration

Bridging classical actuarial science with machine learning for risk prediction.

Read full article →