Build a Telegram AI assistant for tasks, calendar and Gmail

See What's Earning in AI Automation Freelancing.
DigiNo helps new AI automation freelancers earn faster by tracking what clients actually pay for.

Busy professionals are drowning in disconnected tools for tasks, emails, and scheduling. This AI automation pulls all of that into a single Telegram chat interface powered by OpenAI, and freelancers who can set it up are sitting on a repeatable service offering.

What This Automation Does

Listens to Telegram messages (text or voice) and routes each request to the right service automatically
Creates, updates, and completes tasks based on natural language instructions sent by the user
Schedules, reschedules, and checks availability on Google Calendar without the user leaving Telegram
Drafts, sends, and retrieves Gmail messages and pulls Google Contacts on demand

Tools Used

Telegram
OpenAI
Gmail
n8n

Where to Get Hired for This Skill

On Contra, top freelancers across this stack have earned 102 combined verified reviews from real client projects.

Source: Contra freelancer search · refreshed 30 May 2026

Start Earning as a Freelancer on Contra

Contra is a commission-free professional network for independents. Browse live AI automation work and keep what you earn.

Join Contra Free →

How To Build It

Wire Telegram as the input channel

A Telegram bot is created via BotFather and connected as the entry point so every message the client sends arrives as a structured payload the workflow can act on immediately.

Connect Gmail and Google Calendar access

OAuth credentials for Gmail and Google Calendar are authorised under the client's Google account so the workflow can read, draft, and send emails and query or write calendar events on their behalf.

Configure OpenAI as the decision-making layer

An OpenAI API key is added and the agent is given a system prompt that defines its persona, the tools it can call, and the context it should retain between messages within a session.

Map each intent to the correct service action

Routing logic is defined so that task-related phrases trigger the task manager, scheduling phrases hit Google Calendar, email phrases invoke Gmail, and expense phrases write to the tracking store, with the agent deciding which branch to execute.

Test end-to-end with real client scenarios

A set of representative messages covering each capability is sent through the live Telegram bot to confirm that responses are accurate, memory persists within a session, and all credentials are functioning without scope errors.

Pitfalls

Google OAuth tokens expire after a period of inactivity and will silently fail on Gmail and Calendar actions unless a refresh token strategy is in place from day one.
OpenAI's function-calling can misroute ambiguous requests, for example a message like ‘cancel lunch' could hit the calendar, the task list, or both, so the system prompt needs explicit priority rules to prevent double actions.
Voice message transcription adds a second API call and a point of failure; if the audio is noisy or the language switches mid-message, the intent classification degrades and the agent returns a generic fallback instead of completing the task.

FAQ

Can I build this without coding?

Yes. The workflow is built entirely in n8n using visual configuration and credential forms. The only text you write is the OpenAI system prompt, which is plain English.

How long does it take?

A first build from scratch typically takes 90 minutes to 2 hours for someone familiar with n8n and Google OAuth. A second or third delivery for a new client drops to under an hour once you have a template saved.

What can I charge?

Pricing is your call based on the scope you agree with each client. A common structure is a one-time setup fee plus a monthly retainer covering maintenance, credential refreshes, and prompt tuning as the client's needs evolve.

Which tool is required vs optional?

Telegram and OpenAI are required since the entire interface and intelligence layer depend on them. Gmail, Google Calendar, and Google Contacts can each be toggled off if a client does not use them, making the workflow modular for different client setups.

This is original DigiNo analysis. The underlying automation pattern is a community workflow template – view the original on n8n.