How to Build an AI Meeting Scheduler

This ai meeting scheduler finds optimal times across multiple calendars and books meetings without the back-and-forth emails. Automated booking that respects everyone's preferences.

What You Need Before Starting

• Python 3.8+
• Google Calendar API credentials
• Flask for webhook endpoints
• SMTP or Slack for notifications

Step 1: Connect Calendar API

Set up access to Google Calendar or Outlook.

from google.oauth2.credentials import Credentials
from googleapiclient.discovery import build

def get_calendar_service(creds_path):
    creds = Credentials.from_authorized_user_file(creds_path)
    return build("calendar", "v3", credentials=creds)

def get_events(service, time_min, time_max):
    events = service.events().list(
        calendarId="primary", timeMin=time_min, timeMax=time_max,
        singleEvents=True, orderBy="startTime"
    ).execute()
    return events.get("items", [])

Step 2: Build the Logic

Implement the core meeting scheduler algorithm.

from datetime import datetime, timedelta

def find_available_slots(events, date, duration_minutes=30):
    busy = [(e["start"]["dateTime"], e["end"]["dateTime"]) for e in events]
    slots = []
    current = datetime.combine(date, datetime.min.time().replace(hour=9))
    end_of_day = current.replace(hour=17)

    while current + timedelta(minutes=duration_minutes) <= end_of_day:
        slot_end = current + timedelta(minutes=duration_minutes)
        is_free = all(slot_end <= b[0] or current >= b[1] for b in busy)
        if is_free:
            slots.append({"start": current.isoformat(), "end": slot_end.isoformat()})
        current += timedelta(minutes=15)
    return slots

Step 3: Handle Notifications

Send confirmations and updates automatically.

def send_calendar_notification(event, recipients, notification_type):
    templates = {
        "confirmed": "Your {event_type} is confirmed for {time}",
        "reminder": "Reminder: {event_type} in {minutes} minutes",
        "cancelled": "{event_type} has been cancelled",
    }
    message = templates[notification_type].format(
        event_type=event["summary"], time=event["start"],
        minutes=event.get("reminder_minutes", 15))

    for recipient in recipients:
        send_email(recipient, f"Calendar: {event['summary']}", message)

Step 4: Store and Track

Keep a log of all scheduling actions.

import sqlite3

def log_scheduling_action(action_type, event_data):
    conn = sqlite3.connect("scheduling.db")
    conn.execute("""INSERT INTO schedule_log
        (action, event_id, details, timestamp)
        VALUES (?, ?, ?, ?)""",
        (action_type, event_data["id"], json.dumps(event_data), datetime.now().isoformat()))
    conn.commit()

Step 5: Automate the Workflow

Set up triggers and cron jobs.

# Run scheduling checks every 15 minutes during business hours
*/15 8-17 * * 1-5 cd /root/scheduler && python check_schedule.py

def run_scheduled_checks():
    service = get_calendar_service("creds.json")
    events = get_events(service, datetime.now().isoformat(), (datetime.now() + timedelta(hours=1)).isoformat())
    for event in events:
        check_and_notify(event)

What to Build Next

Add meeting type detection. Route discovery calls to 30-min slots, demos to 60-min.

Related Reading

• Building Automated Task Lists from Meeting Transcripts - automated task lists meeting transcripts
• Creating an Automated Meeting Scheduling System - automated meeting scheduling system guide
• Competitive Intelligence with AI - competitive intelligence ai automated