Switching context between your agents has a cost. Humanload will measure it.

Every decision you make, every project you reopen, every session left waiting: finally counted, finally visible. The work nobody sees, and what it takes from you.

Confirm your address by email, and we’ll write to you at launch. Unsubscribe at any time.

Three layers, nothing to fill in

fictional preview of the upcoming product

The load · what it takes from you

  • 23 decisions made
  • 7 project switches
  • 4 sessions waiting

Reference point
5 project switches before noon: your highest this month.

Alongside these counts, you can add your own end-of-day rating. It’s optional. Counts only: no score, no predictions.

The journal · your work, made visible

  • billing12 sessions
  • client-api9 sessions
  • website5 sessions

The journal will show activity by project and the decisions you made during your sessions.

Your retro · what could be done better · opt-in

  • You approved the same command 30 times: add it to your permissions.
  • Your sessions wait twice as long in the afternoon as in the morning. Batch your approvals.
  • PR #214 has been waiting for your review since Thursday.

It looks at how you use AI, not at you. You’ll decide whether to apply its suggestions.

What counts as load

A decision made
An agent proposes a plan or asks permission to run a command. You decide.
A project switch
You leave one repository for another, long enough to answer an agent. Then you come back.
A session waiting
An agent has finished, or is waiting for your answer, while you’re busy elsewhere.

Nothing to fill in: collection will be automatic. The only optional input is your own end-of-day rating, in one keystroke.

Who sees what

A commitment for the upcoming product. Your summary is yours first; you share it only if you choose to.

What is measuredYouYour team and its manager
Your loadYesAggregated with others, with your consent
Your journalYesNo
Your retroYes, if you turn it onNo
The text of your conversationsYesNever

14%

of US workers surveyed who use AI say they have experienced “brain fry”. Those most at risk juggle several tools or oversee several agents.

BCG, reported by Axios, March 6, 2026

“Developers are having to make decisions more often.”

Stack Overflow, May 21, 2026

Our commitments

For the upcoming product.

  1. Nothing is analyzed without your consent: your retro is opt-in.
  2. Load means counts: no score, no predictions.
  3. A team only sees aggregates: never an individual, never any text.
  4. You always know where your data is processed.
  5. You can delete everything, whenever you want.

What the list keeps about you

An example of a confirmed sign-up in the upcoming waitlist, whose database will be stored in the EU:

id
a random identifier
email
you@example.com
lang
en
domain
humanload.app
ref
an approved referral code, if any
status
confirmed
signed_up_at
2026-10-08
confirmed_at
2026-10-08
last_sent_at
2026-10-08
token_hash
null
token_expires_at
null
next_token_hash
null
next_token_expires_at
null

Before confirmation, token_hash stores the SHA-256 hash of the confirmation token, and token_expires_at its expiry. During a resend, next_token_hash and next_token_expires_at store those of the new token. No token is stored in plain text. Without confirmation, the row is deleted 7 days after the latest request.

The list keeps no IP address and no user agent. Unsubscribing deletes the row.

Questions

Will I have to fill anything in?

No. Collection will be automatic. You can add an end-of-day rating in one keystroke, if you want to.

Will my employer see my numbers?

No. A team version will only show aggregates, with your consent. Before any team use, talk to your data protection officer and your employee representatives.

Will conversation analysis be optional?

Yes. It will be part of your retro, which you choose to turn on or not. Before launch, we’ll explain where the analysis will run.

What about the EU AI Act, or workplace risk assessments?

We don’t give legal advice: your data protection officer and your employee representatives are the right people to ask. Humanload won’t score or rank people. Load will be shown as event counts.

When will Humanload be available?

We haven’t announced a launch date yet. We’ll email you when it’s available.

Get an email when Humanload launches.

Join the waitlist

For engineering teams

For a team load assessment based on aggregate data, email hello@humanload.app