On-Device AI Agent for Android Automation
Trending~1 monthAn AI-driven app that controls Android devices, providing on-device automation without cloud reliance.
This project involves building an on-device AI agent for Android devices, similar to PokeClaw, that allows users to automate tasks and control their phones through natural language interactions. The AI agent will run locally, ensuring data privacy and offering a fast, efficient alternative to cloud-based solutions.
MonetizationProvide a freemium model with basic automation features available for free and advanced features, such as custom automation scripts and integration with third-party apps, available through a subscription.
Stack:KotlinTensorFlow LiteAndroid SDKNatural Language Processing
Self-Hosted Notion + AI
~1 monthA Notion alternative with built-in AI writing, search, and automation — all your data stays on your server.
Fork or extend AppFlowy/AFFiNE with an AI layer: AI writing assistant in every document, semantic search across all notes, automatic tagging and linking of related pages, meeting notes → action items extraction. Deploy via Docker. Use local LLM (Ollama) or API. Privacy-first: nothing leaves the server.
MonetizationManaged hosting: $12/user/mo. Enterprise: $49/user/mo with SSO.
Stack:RustTypeScriptReactOllama
Open Source Cursor IDE
~1 monthA VS Code fork with built-in AI coding — uses any model (Claude, GPT, local) with no subscription lock-in.
Fork VS Code or build a VS Code extension that adds: inline code completion (using any OpenAI-compatible API), chat panel for code questions, diff-based edit suggestions, and codebase-aware context via RAG. The value: users bring their own API key and aren't locked into $20/mo. Support local models via Ollama for fully offline use.
MonetizationOpen source core. Pro: $5/mo for cloud context sync + team features.
Stack:TypeScriptVS Code Extension APIAny LLM API
Self-Hosted Perplexity
~1 weekAn AI search engine you own — searches the web, reads the results, and gives you a sourced answer.
Build an AI-powered search: user asks a question → search the web (SearXNG or Brave API) → fetch top 5 results → read and extract key info → synthesize an answer with inline citations. Deploy with Docker. Use any LLM (local via Ollama or API). The key differentiator vs Perplexity: self-hosted, private, no usage limits.
MonetizationSelf-hosted: free/open source. Managed hosting: $19/mo.
Stack:TypeScriptNext.jsSearXNGOllama