h|ve Platform calculator

Nine answers, and you see the size of the gain.

None of them asks for an internal study: the rest is inferred from market averages. The output is a range, never an exact number, because the input is declared by you and not measured by the platform. At the end, the tier the answers call for and how long it takes to pay for itself.

Calculator

How much h|ve can unlock in your engineering, per year.

Nine answers, and none of them asks for an internal study. Everything else is inferred from market averages. The output is a range, never an exact number, because the input is declared by you and not measured by the platform. Where the biggest gain sits changes with the answers: teams barely using AI gain in acceleration, teams already deep in it gain in avoided rework and in policy.

Development team*people
#
Average monthly cost / person*fully loaded
US$

Salary, payroll charges, benefits, tools and infrastructure. The market average for a mid and senior team in Brazil sits close to US$ 3,462 per month.

AI adoption in the team0%

How much of the team uses a coding assistant or agent as routine, not as an isolated experiment.

Monthly AI spend
US$

Seats, APIs and agent runs. A market reference is US$ 46 per assisted person per month. If the shop runs agents at scale, the real number is far higher.

the five that change the math

Each one opens a different line of gain. There is no right answer: there is leverage in different places.

Can a change made with AI be reconstructed?

Requirement, prompt, agent, diff, review and release linked without anyone digging through history.

Is AI spend attributed?

From the vendor invoice down to team, project, repository, agent and delivered outcome.

Time spent looking for context

Requirement, architecture decision, business rule, owner, dependency and policy. High is 7.5 hours per person per week, medium 5.8, low 3.5.

Does what an agent writes pass a gate with an owner before production?

A gate is a block with a named person answering for the crossing, not a notice in the team channel.

How much of the work starts from a written, approved requirement?

With no approved spec there is nothing to compare what the AI delivered against, and the rework only shows up after review.

POTENTIAL ANNUAL GAIN WITH h|ve
—
Answers missing

Fill in the fields marked with an asterisk on the left and the math runs right away. The five maturity questions already carry a middle answer, and you adjust whatever differs in your shop.

Plan and return

Which tier your answers ask for, and how long it takes to pay for itself.

The tier comes from team size and the five maturity questions, and the price is the same one in the grid above, with no parallel table. The return is calculated with the gain ramping in over the first six months, because nobody harvests everything in month one.

As soon as the team size lands in the calculator above, the recommended tier and the monthly bill show up here.
Next step

Not sure which step you are on?

A short conversation with the people who build the platform settles it in fifteen minutes, with your repository and PR numbers on the table.