AI Developer Needed to Build a Voice-Controlled Personal Assistant

I am looking for an experienced developer to build a complete AI personal assistant from scratch. The assistant should be controlled mainly through **voice messages and text messages sent via WhatsApp or Telegram**. The goal is to complete everyday administrative and business tasks that would normally require many clicks and several hours of manual work, simply by sending a voice command. Example: “Find the document called Marketing Plan, summarize it, extract the main tasks and add them to my task list.” The assistant should understand the command, perform the requested actions and send a clear confirmation through WhatsApp or Telegram. I am open to the developer recommending either WhatsApp or Telegram for the first version. The priority is to create the system in the **simplest, most reliable and cost-effective way possible**. ## Main Features The assistant must be able to: * receive text messages; * receive voice messages; * download and transcribe audio messages; * understand natural voice commands; * generate summaries; * create and organize notes; * extract tasks, deadlines and important information; * execute real actions through connected applications; * send results and confirmations back to the user; * maintain conversation context and action history. ## Voice Command Examples The system should understand commands such as: * “Create a note from this voice message.” * “Summarize this audio.” * “Turn this message into a task list.” * “Add this information to my Google Doc.” * “Add a new row to my client spreadsheet.” * “Find the file related to this project and summarize it.” * “Check my calendar and find a free time tomorrow afternoon.” * “Schedule a meeting for Friday at 3 PM.” * “Summarize my important emails from today.” * “Create a draft reply to this email.” * “Show me the tasks I still need to complete.” * “Prepare a summary of my day.” ## Messaging Platform The assistant should work through: * WhatsApp; or * Telegram. The developer should recommend the best option for the MVP based on: * development cost; * monthly operating cost; * setup complexity; * reliability; * support for voice messages and files; * ease of maintenance; * future scalability. The architecture should be designed so that the second messaging platform can be added later without rebuilding the entire system. ## Google Workspace Integration The assistant must securely connect to Google Workspace through OAuth. ### Google Drive The assistant should be able to: * search for files and folders; * find files by name or content; * read supported documents; * create and organize folders; * retrieve file links; * summarize files. ### Google Docs The assistant should be able to: * create new documents; * read existing documents; * summarize documents; * add notes to existing documents; * append new sections; * update specific content; * create structured documents from voice messages. ### Google Sheets The assistant should be able to: * read spreadsheet data; * add new rows; * update cells; * create simple tables; * add clients, notes, expenses, tasks or other information; * search and filter existing data. ### Google Calendar The assistant should be able to: * read upcoming events; * check availability; * identify scheduling conflicts; * create events; * modify events; * cancel events after confirmation; * add attendees and meeting information. ### Gmail The assistant should be able to: * search emails; * read and summarize messages; * identify important or urgent emails; * extract tasks from emails; * prepare draft replies; * send emails only after user confirmation. ## Notes and Task Management The system should include a simple database for notes and tasks. Each note should include: * title; * content; * category; * creation date; * last update date; * source, such as voice, text, email or Google Drive; * optional link to a Google document. Each task should include: * title; * description; * priority; * deadline; * status; * related project; * source; * creation date. The assistant should be able to create, search, update, complete and summarize notes and tasks. ## AI Agent Requirements The assistant must understand the user’s intent and select the correct action. It should distinguish between: * normal questions; * note creation; * summaries; * task creation; * Google Drive searches; * Google Docs updates; * Google Sheets updates; * calendar requests; * email requests. The assistant must never claim that an action has been completed unless the relevant API confirms that it was successful. ## Confirmations and Security Read-only actions may be completed automatically. The assistant must request confirmation before: * sending an email; * deleting a file; * cancelling an event; * sharing a document; * replacing existing document content; * modifying important spreadsheet data; * performing any irreversible action. API keys, OAuth tokens and credentials must be stored securely on the backend and must never be exposed in the frontend. Only authorized user accounts or phone numbers should be allowed to give commands to the assistant. ## Technical Approach I am not committed to a specific technology stack. The developer should recommend the most efficient and affordable architecture. Possible technologies may include: * Telegram Bot API; * Meta WhatsApp Cloud API; * Twilio WhatsApp; * n8n; * Make; * Supabase; * PostgreSQL; * OpenAI API; * Google Workspace APIs; * serverless functions; * Node.js; * TypeScript; * Python. No-code or low-code tools may be used where appropriate, but the final assistant must be fully functional and not only a visual prototype. The priority is: * low development cost; * low monthly operating cost; * reliability; * simple maintenance; * secure architecture; * minimal vendor lock-in; * easy future expansion. ## Expected Workflow 1. The user sends a text message or voice note through WhatsApp or Telegram. 2. The system receives the message. 3. Voice messages are downloaded and transcribed. 4. The AI identifies the user’s intent. 5. The appropriate tool or integration is selected. 6. The requested action is executed. 7. The result is saved in the database. 8. The assistant sends a confirmation or response back to the user. 9. Important actions are stored in an activity log. ## MVP Scope The first working version should include: * WhatsApp or Telegram integration; * text message support; * voice message support; * audio transcription; * AI command interpretation; * note creation and management; * summaries; * basic task management; * Google Drive integration; * Google Docs integration; * Google Sheets integration; * Google Calendar integration; * secure user authentication; * action history; * error handling; * confirmation system for sensitive actions. Gmail integration may be included in the MVP or added as a second phase, depending on cost and development time. ## Deliverables The selected developer will be responsible for: * designing the technical architecture; * building the complete system from scratch; * configuring the messaging platform; * implementing voice transcription; * implementing the AI assistant logic; * connecting Google Workspace; * creating the database; * implementing authentication and security; * deploying the system; * testing all main workflows; * fixing errors discovered during testing; * providing technical documentation; * providing instructions for managing credentials and integrations; * providing a short video demonstrating the completed assistant. The final delivery must include a real working assistant that can receive a voice message, understand the command, perform an action and return the result. ## Developer Requirements Please apply only if you have experience with several of the following: * AI agents; * Telegram Bot API; * WhatsApp Cloud API or Twilio; * voice message processing; * speech-to-text APIs; * OpenAI API; * Google OAuth; * Google Drive API; * Google Docs API; * Google Sheets API; * Google Calendar API; * Gmail API; * n8n or Make; * Supabase or PostgreSQL; * backend development; * API integrations; * secure credential management. Experience building personal assistants, automation systems or AI agents is strongly preferred. ## Application Questions Please answer the following questions: 1. Have you previously built an AI assistant controlled through WhatsApp or Telegram? 2. Can you share examples of similar projects? 3. Which platform would you recommend for the MVP: WhatsApp or Telegram, and why? 4. What technical stack would you use? 5. How would you process and transcribe voice messages? 6. How would you connect the assistant to Google Drive, Docs, Sheets and Calendar? 7. How would you protect personal data and API credentials? 8. What limitations should I be aware of? 9. What is your estimated development time? 10. What is your fixed-price quote for the MVP? 11. What would the estimated monthly operating cost be? 12. What information and account access would you need from me? ## Budget and Contract I prefer a **fixed-price contract for the complete MVP**. Please provide: * total development price; * estimated delivery time; * estimated monthly operating costs; * list of included features; * list of any features that would require a second phase. I am looking for a developer who can take responsibility for the entire project with minimal technical involvement required from me.