The Post-App Phone is Here
We spend our lives tapping through a grid of apps. But what if a single command could orchestrate them all? Meet Konduct, an AI shell that turns your intent into action.
⚡ The Signal
The smartphone interface is frozen in time. For fifteen years, we’ve been tapping on a grid of app icons—a digital filing cabinet for siloed experiences. But the hardware and the AI are finally powerful enough to ask a bigger question: what if your phone just did what you wanted? With whispers that OpenAI is developing its own app-less device, the race is on to build the post-app interface.
🚧 The Problem
We think in outcomes, but our phones force us to think in apps. To plan a simple trip, you open your calendar, then a flight app, then a hotel app, then a map app, then a messaging app to share the details. You are the human API, manually stitching together data from five different services. This app-centric model is inefficient, cluttered, and fundamentally misaligned with how people solve problems. The phone should be a partner that understands your intent, not a switchboard operator that just connects you to the right app.
🚀 The Solution
Enter Konduct, an AI-native launcher for Android that replaces your home screen with a simple, conversational interface. Instead of navigating a grid of icons, you state your intent in natural language: "Book me a flight to SFO next Tuesday morning, find a hotel near the conference center, and add it all to my calendar." Konduct parses the complex request and orchestrates the apps you already have installed—United, Hilton, Google Calendar—to execute the entire workflow from a single command. It’s not a new phone; it’s a new way of using the one in your pocket.
🎧 Audio Edition
Listen to Ada and Charles discuss today's business idea.
If you're reading this in your email, you may need to open the post in a browser to see the audio player.
💰 The Business Case
Revenue Model
Konduct will operate on a freemium model. The free tier offers a limited number of multi-app actions per month, giving users a powerful taste of the experience. A Pro subscription, priced at $5-10/month, will unlock unlimited actions, priority processing for complex tasks, and advanced features for building and saving custom workflows.
Go-To-Market
The strategy is developer-first and community-driven. We'll start by open-sourcing the core app-control engine to attract a community of power users. To capture organic interest, we will build a public library of 'Automated Workflows' that solve common problems, turning long-tail search queries into user acquisition. A launch on Product Hunt will be paired with a feature allowing users to share their own "recipes," creating a network effect as the community builds and distributes its own custom automations.
⚔️ The Moat
While Konduct competes with incumbents like Google Assistant and hardware plays like the Rabbit R1, its unfair advantage is the data it accumulates on user intent and preferences. Every successfully completed workflow ("user prefers aisle seats on United and nearly always books a Hilton") refines our proprietary dataset. This allows Konduct to fulfill future requests with greater speed and accuracy than generic assistants, creating a powerful data moat that improves with every user action.
⏳ Why Now
The market is at a tipping point. Consumer demand for AI-first experiences is surging, and investors are taking notice, with concepts like an AI home screen for the iPhone already raising funds. More importantly, the regulatory environment is shifting. In a move that could break the incumbents' stranglehold, Europe is considering forcing Google to open Android to third-party AI assistants. This creates the perfect opening for a software-only solution like Konduct to deliver a next-gen experience on today's hardware, today.
🛠️ Builder's Corner
This is just one way to build it, but here's a recommended MVP stack. The user-facing app is an Android Launcher built with React Native (using the Expo framework for speed). User commands (voice or text) are sent to a Python backend running FastAPI. The backend uses a large language model (LLM) to understand the user's intent and translates it into a structured JSON object representing a sequence of actions. This JSON is sent back to the device, where the React Native app uses Android’s Accessibility APIs and app deep-linking to sequentially open and interact with the necessary apps, executing the workflow in the foreground.
Legal Disclaimer: GammaVibe is provided for inspiration only. The ideas and names suggested have not been vetted for viability, legality, or intellectual property infringement (including patents and trademarks). This is not financial or legal advice. Always perform your own due diligence and clearance searches before executing on any concept.
