Del is a personal AI executive assistant designed to act as a proactive, 24/7 digital aide. Created to help users stay organized and accountable, the assistant operates through direct communication, removing the need for a dedicated app. It functions by engaging users directly via text, allowing for a seamless integration into daily workflows without requiring the user to constantly open an interface to manage tasks or review schedules. The service is built upon robust privacy principles, ensuring that data is handled with care and that AI models are not trained on user-specific content.

Functionality of the service revolves around intelligent automation and personal task management. It acts as a bridge between a user's calendar, notes, and tasks, synthesizing this information to provide proactive reminders and assistance. By analyzing existing schedules and commitments, Del can identify potential conflicts or missed opportunities and offer solutions or draft content on the user's behalf. The assistant is designed to be conversational and helpful, often prompting the user to act on pending items or providing drafts for upcoming meetings and correspondence.

Some of the key features are:

• Proactive Task Management: The assistant monitors your to-do lists and schedules, offering reminders and checking in on your progress.
• Calendar Integration: Connects with Google Calendar to display, create, modify, or delete events directly through natural language requests.
• AI-Powered Assistance: Utilizes Google Gemini large language models to generate suggestions, draft documents, and compose SMS replies based on your specific context.
• App-less Interaction: Operates primarily via SMS, allowing users to interact with their assistant in the same way they communicate with colleagues and friends.
• Mac Integration: Features a Mac bridge that reads local Messages database content to provide better context and assistance.
• Privacy-First AI: Operates under terms that prohibit the use of personal user data for training general-purpose AI models.

Operationally, the service functions as a messaging partner. Users authenticate their accounts to start, and upon linking integrations like Google Calendar or phone numbers for SMS, the assistant begins to ingest relevant metadata to build a context-rich profile. When a user interacts with the assistant via text, the system processes these requests against the stored tasks, calendar events, and notes. Responses are generated in real-time, and actions—such as updating a calendar entry or drafting a note—are executed immediately to reflect the user's intent. All interactions are logged, allowing users to review the assistant's performance and the actions taken on their behalf.

Some common use cases include:

• Meeting Preparation: Asking the assistant to draft a document for an upcoming meeting based on calendar context and previous notes.
• Schedule Adjustment: Quickly rescheduling appointments or clearing time blocks via a simple text message when plans change unexpectedly.
• Accountability Reminders: Receiving nudges when tasks are left incomplete or when deadlines are approaching for high-priority projects.
• Note Organization: Capturing quick thoughts or task reminders on the go without needing to launch a complex note-taking application.