I’m a morning person, on paper. I wake up early, I have the calendar, I have the intention. And then somehow by 11am I’ve done two standups, answered twelve Slack messages, and mentally drafted a third one. Zero deep work done.

This is embarrassing to admit. I’m building a productivity app because I personally can’t get this right.

What Pomodoro gets wrong

I tried Pomodoro. For a while it worked. Then meetings got added, then Slack got noisier, then the 25-minute blocks started feeling arbitrary. The problem with Pomodoro isn’t the technique. It’s that it treats every day like a blank canvas. Real calendars are not blank canvases. Tuesday at 2pm after a 90-minute strategy call is not the same as Thursday morning after a good night’s sleep. Scheduling a deep block without knowing what cognitive resources you actually have is guessing.

The hypothesis

There are two signals that actually tell you when to go deep:

1. Your calendar. What’s already happening, when are the gaps, how long are they really?
2. Your HRV (heart rate variability). A proxy for how recovered your nervous system is. Low HRV after poor sleep means your pre-frontal cortex is running at 60%. No amount of willpower fixes that; you’re better off doing shallow work and recovering.

The idea behind Flow is simple: give it calendar access + an HRV reading from your wearable, and it uses Claude (Anthropic’s AI) to surface 2 to 3 realistic deep work blocks for your day. Not aspirational slots. Real ones. And when your focus score is low, it tells you to rest instead of pushing through.

What exists right now

I’m in build-in-public mode so I’ll be straight:

• The Android app exists. It generates plans and runs focus sessions. The AI coaching layer (Claude Haiku 4.5) is wired in the backend. I’m testing it on my own device daily.
• HRV integration is in progress. Android Health Connect is stubbed, real data is next.
• Launch target is Q3 2026. There’s a founding-member tier at $29.99/year (one payment, two-year price lock) for people who join the waitlist now and convert at launch.
• Nothing is live in the Play Store yet. You’re early.

I’m building this because the problem is real and I’m not going to wait until I have a polished story to tell it. The messy version is closer to the truth than the polished version anyway.

What I want from you

If you’ve made it here, join the waitlist at flowdeep.app. And if the HRV + burnout-detection angle interests or worries you, reply to the welcome email or find me on X. I’m reading everything.

Sandi

Follow along: @useflowdeep on X.