{
  "id": 14787316,
  "title": "Turning WhatsApp Bursts into Briefs: A Local AI Tool for Freelancers & Small Businesses",
  "url": "https://urgent.news/2026/10/15/turning-whatsapp-bursts-into-briefs-a-local-ai-tool-for-freelancers",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-15T20:51:40.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/mritunjai/turning-whatsapp-bursts-into-briefs-a-local-ai-tool-for-freelancers-small-businesses-4oi4"
  },
  "original_language": "en",
  "account": "In the realm of freelance project management, communication often becomes a chaotic stream of messages on WhatsApp. Freelancers and agency owners rarely receive clear project briefs; instead, clients flood them with sporadic fragments containing shifting scopes, budgets, and deadlines. Mritunjaii's solution to this problem is Client Buddy, an AI-powered tool designed to streamline this communication process.\n\nClient Buddy is a privacy-first, locally hosted client intelligence dashboard that securely captures live WhatsApp message streams using authenticated headless browser sessions. It then aggregates these messages using a custom 2-minute sliding-window batching engine, grouping bursts of messages together before processing them entirely on the user's device.\n\nOnce a client finishes typing, Client Buddy analyzes the grouped messages and generates structured executive briefs, detailing explicit deliverables, hard deadlines, and tone-matched suggested reply drafts. This local processing ensures that sensitive conversation histories never leave the user's machine, offering enterprise-grade intelligence without exposing data to cloud LLM services.\n\nTo maintain professional boundaries, the application enforces a strict Zero Auto-Reply policy. It never sends messages to clients automatically, ensuring all outbound communication requires human discretion through one-click copy-paste drafts. Moreover, in case of severe incidents such as server crashes or payment failures, the system flags the batch as CRITICAL and triggers an immediate outbound phone call alert to the user's mobile device, using synthesized AI voice for clarity.\n\nThe project is built as a modern monorepo, separating the user-friendly dashboard from the background workers handling the heavy data processing. The Ingestion Engine uses whatsapp-web.js with Puppeteer to securely capture live incoming messages, while the Sliding-Window Batcher prevents disjointed summaries by resetting a 120-second inactivity countdown for each chat. MongoDB stores these pending batches, employing atomic lease locks to guarantee thread safety and exactly-once processing across background workers.\n\nClient Buddy runs locally using Ollama and open-weight models like llama3.2:latest or gemma:2b. The background worker feeds these aggregated batches into the open-weight models, which are then prompted to output structured JSON data that populates the Next.js dashboard with requirements, open questions, and drafted replies. If the local LLM classifies a burst as a CRITICAL emergency, the backend calls the ElevenLabs API to synthesize a natural-sounding voice briefing and uses Twilio to ring the user's mobile phone with the alert.\n\nPrivacy is at the core of Client Buddy's design. Raw source messages are protected by a native MongoDB 7-day TTL index, automatically purging these records after seven days to comply with GDPR regulations. The use of open innovation, specifically local inference through Ollama and open-weight models, ensures that sensitive conversation histories remain within the user's machine for analysis. This approach provides advanced AI capabilities without compromising client trust or incurring per-token costs associated with cloud LLM services.\n\nMritunjaii submitted Client Buddy for several prize categories, including Best Use of Gemma for utilizing the open-weight model as the core reasoning engine, Best Use of MongoDB Atlas for its role in complex state management and privacy-first data handling, and Best Use of ElevenLabs for the synthesized alerts. This tool represents a significant leap forward in managing client relationships while prioritizing data privacy and security in the freelance and small business sectors.",
  "summary": "This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built Client communication on WhatsApp is notoriously chaotic. Freelancers and agency owners rarely receive structured project briefs; instead, clients send rapid-fire bursts of fragmented messages that bury essential scope shifts, budgets, and deadlines. I built Client Buddy (formerly ClientPulse AI),…",
  "key_points": [],
  "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."
}