SELECTED WORK
Three projects. Research, data science, and a live AI product.
Three projects. Research, data science, and a live AI product.
AI Audit Pipeline
Built and shipped a fully automated AI audit pipeline in one week, solo. Form intake to GPT-4o generation to email delivery, no human in the loop. Built for mission-driven founders who can't afford a $5k consulting sprint.
01 · AI Product · Shipped
Ethic Flow:
AI Audit Pipeline
Built and shipped a fully automated AI audit pipeline in one week, solo. Form intake to GPT-4o generation to email delivery, no human in the loop. Built for mission-driven founders who can't afford a $5k consulting sprint.
OPENAI API
SHIPPED PRODUCT
PYTHON
01 · AI Product · Shipped
Ethic Flow:
AI Audit Pipeline
Built and shipped a fully automated AI audit pipeline in one week, solo. Form intake to GPT-4o generation to email delivery, no human in the loop. Built for mission-driven founders who can't afford a $5k consulting sprint.
OPENAI API
SHIPPED PRODUCT
PYTHON
SentimentAnalysis & Behavioral Modeling
Can behavioral patterns predict human decision making? Scraped 450+ Reddit posts, built an NLP pipeline with 85% predictive accuracy, and prototyped a bias awareness feature to surface unconscious patterns before decisions are made.
02 · NLP & Predictive Modeling
Decoding Desire:
SentimentAnalysis & Behavioral Modeling
Can behavioral patterns predict human decision making? Scraped 450+ Reddit posts, built an NLP pipeline with 85% predictive accuracy, and prototyped a bias awareness feature to surface unconscious patterns before decisions are made.
npl
SCIKIT-LEARN
PYTHON
02 · NLP & Predictive Modeling
Decoding Desire:
SentimentAnalysis & Behavioral Modeling
Can behavioral patterns predict human decision making? Scraped 450+ Reddit posts, built an NLP pipeline with 85% predictive accuracy, and prototyped a bias awareness feature to surface unconscious patterns before decisions are made.
npl
SCIKIT-LEARN
PYTHON
& Churn Analysis
A solo mixed-methods investigation into why users were leaving. Backend data analysis, surveys, and depth interviews surfaced a 59.6% churn rate and a median session count of 1. Findings were adopted by stakeholders and drove a full platform redesign.
03 · Strategy & Analytics
Poolhouse: Product Strategy
& Churn Analysis
A solo mixed-methods investigation into why users were leaving. Backend data analysis, surveys, and depth interviews surfaced a 59.6% churn rate and a median session count of 1. Findings were adopted by stakeholders and drove a full platform redesign.
CHURN ANALYSIS
DATA ANALYTICS
USER INTERVIEWS
03 · Strategy & Analytics
Poolhouse: Product Strategy
& Churn Analysis
A solo mixed-methods investigation into why users were leaving. Backend data analysis, surveys, and depth interviews surfaced a 59.6% churn rate and a median session count of 1. Findings were adopted by stakeholders and drove a full platform redesign.
CHURN ANALYSIS
DATA ANALYTICS
USER INTERVIEWS
More projects coming soon
More projects coming soon
· · ·
Let's work on something
that matters.
Let's work on something
that matters.