Before
Jordin's calendar, to-do lists and notes lived in separate places, each needing its own attention. They wanted one assistant that could listen, understand the request and act across all of them. Getting messages to flow cleanly between platforms was the hard part, with filters that misfired along the way.
“Built a multi-functional voice agent capable of managing calendars, tasks, and data storage.”
In their own words
The turning point
In the program, Jordin leaned into building with AI, using ChatGPT, Claude and Make.com's own AI to shape and improve the workflows. They chose a modular design, sorting every request into Productivity, General or Memory and branching each into its own path. When tasks were unclear or failed, they built feedback loops instead of giving up.
After
Jordin now has OmniTask, a voice agent that answers through Telegram and is designed to work with Discord, WhatsApp and Slack. It checks and books appointments with email notices for attendees, manages Notion to-do lists and saves transcribed messages and task records in Airtable. Next on their list is a research agent that tracks trends for business strategy.
What it lets people do
OmniTask lets someone speak or type a request in a chat app and have it handled across their tools. Meetings get booked, tasks get filed, and conversations are remembered in one central place, with context carried from message to message.
What they learned
Jordin learned that automation is built through iteration: test, see what breaks, fix it and try again. They also found that bringing AI into the loop makes task management feel personal, and that handling voice and text from many channels is manageable when the design stays modular.
“Iterative problem-solving for automation challenges.”
In their own words
OmniTask Voice Agent was built during AI Mastery.
Read the full submission
[JORDIN iLAC]
* NOTE: not sure if this a duplicate, but it was supposed to go into this category. sorry about that
Innovative Project Name
"OmniTask Voice Agent"
Image Analysis
Chosen Category:
"The Automation Architect"
Explanation:
- The project aligns with "The Automation Architect" category due to its visionary approach in building AI-driven automations. By integrating platforms like Telegram, Airtable, and Make.com, the project establishes fully automated workflows that streamline personal and business productivity tasks.
- The modular design and scalability further emphasize its role in leveraging APIs and AI tools for task management and data organization.
The emphasis on adaptability, task categorization, and future-proofing through additional agents also supports the choice of this category.
Comprehensive Summary
Highlights:
- Voice Agent Functionality:
- Activated through various platforms like Telegram, Discord, WhatsApp, and Slack.
- Supports both text and voice inputs for user interaction.
- Categorizes tasks into Productivity, General, and Memory.
- Employs conversational context for more intuitive task execution.
- Core Features:
- Calendar management: Check and book appointments with email notifications for attendees.
- To-do list integration: Sync with Notion and manage task lists.
- Data storage: Stores transcribed messages and task records in Airtable.
- Aggregation: Combines and organizes data using a sorting director and text aggregator.
- Research agent: Tracks trends and relevant data for business purposes.
- Tools Used:
- Telegram Desktop: Primary interaction platform.
- Notion: Task and to-do list management.
- Airtable: Centralized storage for conversations, tasks, and memory data.
- Make.com: Automation tool for linking workflows.
- Methodology:
- Modular, scalable design with visual representations via Make.com.
- Data organization through arrays and aggregators.
- Workflow branching based on categories such as productivity, research, and task management.
- Challenges and Learnings:
- Filter issues in message processing.
- Iterative design requiring feedback loops for unclear or failed tasks.
- Development of multi-level bots for scalability.
- Leveraging AI platforms (ChatGPT, Claude, and Make.com AI) to optimize workflows.
- Future Plans:
- Expand automation in Airtable to reduce costs and enhance efficiency.
- Develop separate scenario pages for individual agents to avoid information overload.
- Implement a Research Agent for trend analysis and strategy alignment.
Project Description
Achievements:
- Built a multi-functional voice agent capable of managing calendars, tasks, and data storage.
- Enabled seamless integration across communication platforms and productivity tools.
Tools Used:
- Telegram Desktop and Mobile, Notion, Airtable, Make.com, ChatGPT, Claude, and RAG System (Retrieval-Augmented Generation).
Methodology:
- Modular design using Make.com's visual modules.
- Aggregation of conversational data through arrays and sorting directors.
- Categorization of tasks into defined workflows for better scalability and usability.
Things Learned:
- Iterative problem-solving for automation challenges.
- Integration of AI for personalized task management.
- Efficiency in handling multi-channel inputs (voice and text).
Relevant URLs (if any):
- Make.com: https://www.make.com/
- Airtable: https://www.airtable.com/
- Notion: https://www.notion.so/
Written 21 January 2025· Submitted by the storyteller
Every story is written by a Mindvalley student or by someone our teachers have worked with. How we collect and check stories
