L / 01 — Open lab

TeamPulse AI Workspace

Google Workspace dashboard that turns emails, calendar, and tasks into live team intelligence — built in an hour.

Gemini AIGCPGoogle Workspace

"Most team dashboards ask you to enter data. TeamPulse reads the signals already in your Google Workspace and extracts the picture for you."

I — Overview

Why it exists.

TeamPulse started as a question: what if Gemini read your inbox, calendar, and task list so your team didn't have to write status updates? Built in roughly an hour for a rapid R&D sprint, it ended up answering that question cleanly.

The platform pulls Google Workspace signals — emails, calendar events, Drive activity, and tasks — and pipes them through Gemini in the data layer. The model extracts tasks, decisions, blockers, and workload summaries without any manual entry. The output is a live workspace dashboard: a structured, scannable view of team health that updates as the signals do.

Proactive nudges are first-class. When Gemini detects a blocker or a deadline slipping, the platform surfaces an alert rather than waiting for someone to notice. The whole thing runs on Google Cloud Services — Cloud Run, Firestore, and the Workspace APIs — so there's no infra to manage outside the GCP console.

II — What it does

The surface.

01

Gemini-powered signal extraction

Emails, calendar events, and tasks flow through Gemini in the pipeline. Tasks, decisions, and blockers surface automatically — no manual tagging.

02

Live team health view

Structured workspace dashboard that reflects the current state of work across the team in real time.

03

Workload summaries

Per-person and per-project summaries generated by the model, not assembled by hand.

04

Proactive nudges

When a blocker or deadline risk is detected, the platform surfaces an alert before it becomes a fire.

05

Full Google Workspace integration

Gmail, Calendar, Drive, and Tasks — the signals your team already produces, read automatically.

III — The stack

Under the hood.

Frontend
Next.js
AI
Gemini AI
Cloud
Google Cloud RunFirestore
Integrations
Google Workspace APIsGmailCalendarDrive
IV — Notes

Worth knowing.

  • Built end-to-end in approximately one hour as a rapid R&D prototype.
  • Runs entirely on Google Cloud Services — no external infra.