Ep. Log

A tiny webapp for the eternal problem: "wait... which episode was I on?"

Track what you're watching, bump the episode counter with one click, and ask an LLM (via Groq) for what to watch next based on your list.

---

Why I built this

I watch a lot of anime and constantly lose track of which show and episode I'm on. This is a small side project to fix exactly that — with a bit of fun added on top: an LLM-powered "what should I watch next" recommendation based on my current list.

Tech stack

Layer	Tech
Backend	Flask (Python)
Frontend	Vanilla HTML/CSS/JS
Storage	data.json (flat file, no DB)
Recommendations	Groq API (openai/gpt-oss-120b)

Features

• Add anime with title, current episode, status, and notes
• One-click +1 to bump your episode count as you watch
• Delete entries you're done with (or embarrassed by)
• Ask for a recommendation — sends your watchlist to Groq and gets back 3 picks with reasons, powered by an open-weight LLM running at Groq's LPU speeds

Setup

git clone https://github.com/SHAROZ221/ep-log.git
cd ep-log
python -m venv venv
source venv/bin/activate   # on Windows: venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env

Open .env and paste your Groq API key:

GROQ_API_KEY=your_key_here

Get a free key at console.groq.com/keys — no credit card required.

Run it

python app.py

Open http://127.0.0.1:5000 in your browser.

Project structure

ep-log/
├── app.py               # Flask backend — CRUD routes + Groq recommend endpoint
├── requirements.txt
├── .env.example
├── data.json             # your watchlist, stored locally
├── templates/
│   └── index.html
└── static/
    ├── style.css          # manga-panel inspired theme
    └── script.js

Notes

• Data is stored locally in data.json — no database needed
• If you don't set a GROQ_API_KEY, everything still works except recommendations
• Built as a side project while pivoting toward GenAI/LLM application engineering — a small, complete example of wiring a Flask app to an LLM API end-to-end