Inspiration
Procrastination feels harmless because its cost is invisible. I wanted to build a system that makes avoidance measurable and cumulative instead of forgotten.
What it does
Delay Ledger generates contextual excuses, tracks real-time wasted minutes, builds a streak and debt system, ranks destructive moments using a Shame Score, and pushes the user into a timed micro-action sprint.
How we built it
Frontend in vanilla HTML, CSS, and JavaScript with localStorage state management. A lightweight Node.js backend communicates with a local LLM via Ollama for contextual excuse and micro-step generation.
Challenges we ran into
- Real-time tracking without artificial increments
- Stable snapshot-based scoring
- Reliable AI output formatting
- Maintaining a consistent state across refresh ## Accomplishments that we're proud of
- Built a complete behavioral feedback loop
- Designed a ranking system (Hell of Shame)
- Integrated fully local AI inference
- Created a psychologically grounded escalation mechanism ## What we learned Designing behavior-driven UX systems and integrating local AI pipelines without relying on cloud APIs. ## What's next for Delay Ledger Long-term behavioral analytics, adaptive escalation logic, and deeper personalization.
Built With
- css3
- express.js
- html5
- llm-(llama-3.2)
- node.js
- ollama
- vanilla-javascript
Log in or sign up for Devpost to join the conversation.