Always-on AI agent inside Google

The AI that clears your inbox

In partnership with

Those AI newsletters stack up in my inbox until a guilty Sunday-night skim clears them out. We've all done it. So when a walkthrough of Google's new Gemini Spark crossed my desk, my first thought was simple: finally, something that skims for me.

The breakdown comes from the creator who scored early access through the Google Ultra plan and ran real demos on real accounts. Spark isn't another chat window: it's a personal agent inside the Gemini app that runs 24/7, even when your laptop and phone are off. The Verge called it the most impressive and terrifying AI experience they had, and after watching the demos, I get why.

The GTM bets that shouldn't have worked, and did

One grew revenue 50x after half his team quit over the strategy. One brought in 50K signups in a single day with no paid budget. One generated 100M+ views from a stunt that took 50 hours to conceive. One asked every prospect to demo the product themselves instead of demoing it for them.

None of them followed the safe playbook. They treated GTM like an experiment, moved before they had proof, and made bets most founders would never get approved.

HubSpot for Startups documented all 6 stories in the free Bold Bets Playbook. The risks they took, why it was risky, and what it returned.

*Ad

An Agent That Lives Inside Your Google World

Spark sits inside the Gemini app and plugs straight into your stack: Gmail, Calendar, Drive, Sheets, even YouTube. Other tools like ChatGPT and Claude can reach into Google, but none of them live inside the ecosystem the way Google's own agent does. That means it watches your inbox, reacts to new Sheets, and pulls insights on a schedule while you're away.

Here's the part that changes the math: it touches your actual accounts. Not a sandbox. When the author asked it to scan his inbox, it read his real newsletters; when he asked it to log a client, it wrote a real row in a real spreadsheet and spun up a real Drive folder.

The Workflow That Sold Me

The creator told Spark to scan four AI newsletters, pull the top three recurring themes, build a Google Doc, and draft three LinkedIn posts. A couple of minutes later it surfaced ideas like the shift into the agent era. The drafts were clean, but they didn't sound like him yet.

So he taught it a skill: read my last 15 emails, build a style guide, save it as my ghostwriter. Skills work like saved instructions, much like Claude's: teach it once, reuse it forever. The rewrite nailed his voice, short paragraphs, plain text, no emojis.

Then he put it on a clock. Every Friday at 3 p.m., scan the week's newsletters and send a recap. He also built an inbox rule: when an email asks about his services, add a row to his client tracker and create a folder named after the sender, all without a single click.

PRDs by voice. Bug reports by voice. Ship faster.

Dictate acceptance criteria and reproductions inside Cursor or Warp. Wispr Flow auto-tags file names, preserves syntax, and gives you paste-ready text in seconds. 4x faster than typing.

*Ad

Where I'd Slow Down Before Connecting Anything

This is experimental, and you're handing it a lot: real access to Gmail, Drive, and Calendar, plus permission to act inside them. The original poster tests it on a secondary account, never his main one, and says you should watch it closely. Smart caution.

I think that restraint is the real lesson here: an agent that can write to your spreadsheet is powerful, and powerful cuts both ways. During setup you can grant only the apps you trust, skip YouTube, keep Calendar. It's Ultra-plan only and rolling out slowly, but it reaches everyone eventually, and if you live in Google Workspace, this is the one to watch.

This is a powerful approach for anyone drowning in inbox chores who wants the busywork handled while they sleep. If you want to see every demo play out on screen, check out the original article.

Credits to the creator behind the Gemini Spark walkthrough.

One AI employee. Engineering, finance, growth, ops.

Last week Viktor opened 14 pull requests, closed two month-end books, drafted a board update, deployed three landing pages, and triaged 600 support tickets. From inside Slack and Microsoft Teams. 20,000+ teams now run this way.

*Ad