{
  "id": 11976654,
  "title": "PromisePocket",
  "url": "https://urgent.news/2026/10/04/promisepocket",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-04T18:07:27.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/sayandas001/promisepocket-4ah"
  },
  "original_language": "en",
  "account": "PromisePocket is a personal commitment management platform designed by the author for their best friend and roommate, Rahul. The app addresses the issue of people struggling to remember small, personal promises due to limited mental bandwidth. Unlike traditional productivity tools, PromisePocket uses open-weight AI to understand natural language multi-clause commitments, detect ambiguities, and ask for user confirmation before saving them as structured, searchable personal memories with reliable reminders. The app's core philosophy is to \"Remember the little things. Keep the promises that matter.\"\n\nWhen a user types or speaks a promise like \"I'll call Ma tomorrow at 7 PM and return Rahul's book on Friday,\" PromisePocket splits it into two separate, structured proposals. Before scheduling any commitment, the user reviews, edits time or category, and can accept or dismiss the suggestion. The conversational memory feature allows users to ask questions like \"What did I promise Rahul?\" or \"What do I have to do this weekend?\" and receive grounded answers based on their actual stored commitments.\n\nPromisePocket's reminder engine delivers reliable notifications in-app or via the browser, even across app restarts. Users can organize their commitments around the people in their lives using the Relationship-Centered Circle feature, which categorizes promises by Family, Friends, Colleagues, or Health. The app's codebase is structured into frontend (React 18, Vite, TypeScript, Tailwind CSS, Lucide icons, and responsive design) and backend (FastAPI, Pydantic V2, Uvicorn, and clean layered architecture) components. It's containerized with a single Dockerfile for easy deployment on cloud platforms. The app is built using Google's open-weight Gemma 2 AI model, with optional cloud inference via Groq or an in-built deterministic engine if no external endpoint is available.",
  "summary": "This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I Built I built PromisePocket for my best friend and roommate, Rahul. Like many college students juggling coursework, exam deadlines, and personal life, Rahul is someone with a huge heart who constantly wants to be there for everyone. He frequently makes genuine, heartfelt promises in casual conversations or…",
  "key_points": [
    "PromisePocket helps users remember small personal promises",
    "App uses open-weight AI to understand natural language commitments",
    "Relationship-Centered Circle organizes commitments by relationships"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}