Adding a Project

Each project is one Markdown file inside _projects.

Fast Steps

1. Copy templates/project.md.
2. Paste it into _projects.
3. Rename it like this:

2026-05-27-my-project-name.md

1. Edit the title, date, field, tools, links, excerpt, and sections.
2. Commit and push. The project will appear automatically on /ai-data/.

Project Fields

Use one of these values so the project appears in the right tab:

field: "AI"
field: "ML"
field: "EDA"
field: "Stats"
field: "Tool"

• AI: LLMs, RAG, agents, generative systems, NLP, computer vision, and projects that feel like intelligent systems.
• ML: prediction models, recommenders, clustering, forecasting, model comparison, and classic machine learning notebooks.
• EDA: exploratory analysis, wrangling, visualization, and data storytelling without a strong modeling focus.
• Stats: A/B testing, experimental design, causal thinking, and statistical study notes.
• Tool: apps, packages, scripts, scrapers, utilities, and things built to help people work.

Good Project File Names

Use lowercase words and hyphens:

2026-05-27-customer-churn-model.md
2026-06-02-llm-study-assistant.md
2026-06-14-sales-dashboard.md

Minimum Project Template

---
title: "Customer Churn Prediction"
date: 2026-05-27
status: "Completed"
field: "ML"
tools:
  - Python
  - Pandas
  - Scikit-learn
github: "https://github.com/..."
demo:
kaggle:
excerpt: "A machine learning model that predicts customer churn and explains the strongest risk signals."
---

## The Short Version

I built...

## Problem

...

## Approach

...

## Result

...

## What I Learned

...