Generative AI Google Brain

gemini-use-cases-productivity-productivity-coaching-with-google-calendar

3 min read

Productivity Coaching with Gemini and Google Calendar

Source notebook

Repo path: gemini/use-cases/productivity/productivity_coaching_with_google_calendar.ipynb · Open on GitHub · intermediate

Uses Gemini 2.5 Flash with Google Calendar data to provide productivity coaching and schedule recommendations.

Summary

This notebook teaches how to use Gemini as a productivity coach with the Google Gen AI SDK on Vertex AI. It first analyzes sample calendar screenshots from Cloud Storage, then registers a Python function backed by the Google Calendar API as a Gemini tool. The end-to-end workflow covers authentication, client setup, system instructions, multimodal image analysis, function calling, and personalized calendar-based coaching.

Key code patterns

Create Vertex AI GenAI client

client = genai.Client(
    vertexai=True,
    project=PROJECT_ID,
    location=LOCATION,
)

Initializes the Google Gen AI SDK client for Vertex AI using project and region settings.

Load image from Cloud Storage

calendar_image_part = Part.from_uri(
    file_uri=image_file_uri,
    mime_type="image/png",
)

Shows how to pass a Cloud Storage image URI into Gemini for multimodal schedule analysis.

Use system instructions

config=GenerateContentConfig(
    system_instruction=system_instruction,
)

Constrains Gemini to act as a productivity coach with explicit analysis criteria.

Register Calendar API function as tool

response = client.models.generate_content(
    model=MODEL_ID,
    contents=prompt,
    config=GenerateContentConfig(
        system_instruction=system_instruction,
        temperature=0.9,
        tools=[get_next_n_calendar_events],
    ),
)

Lets Gemini call a Python function that retrieves live Google Calendar events.

Retrieve upcoming calendar events

service = build("calendar", "v3")
events_result = service.events().list(
    calendarId="primary",
    timeMin=now,
    maxResults=num_of_events,
    singleEvents=True,
    orderBy="startTime",
).execute()

Uses the Google API Python client to fetch ordered upcoming events from the primary calendar.

Models & APIs used

  • Models: gemini-2.5-flash
  • APIs / services: Vertex AI, Google Calendar API, Cloud Storage
  • SDKs / libraries: google-genai, google-api-python-client

When to use this

Use this pattern when building a Gemini assistant that combines multimodal schedule analysis with live Google Workspace data through function calling.

Gotchas & caveats

  • Requires an existing Google Cloud project with the Vertex AI API enabled.
  • Google Calendar API must be enabled separately.
  • Colab requires explicit user authentication.
  • Calendar access uses ADC with cloud-platform and calendar.events.readonly scopes.
  • Calendar credentials are stored in memory for the current session and require re-authorization after runtime termination.
  • Google Workspace APIs have quotas, and specifying the project applies the correct quotas.
  • get_next_n_calendar_events only accepts values from 1 to 250.

Best practices

  • Use system instructions to define Gemini’s coaching role and analysis criteria.
  • Start with sample calendar screenshots before connecting live user data.
  • Request only readonly Calendar API access for event analysis.
  • Validate tool input bounds before calling the Calendar API.
  • Return a concise subset of event fields: summary, start, end, and status.
  • Handle Calendar API HttpError and return an empty list on failure.

Graph View

Backlinks