Where the money leaks
Eight common leaks in an AI bill. Each fix keeps the answers as good.
Signs your AI bill leaks
If you see one, the bill has room to fall. If nothing changes, it grows with every new customer.
- The bill grows faster than your customers.
- Nobody can say what one answer costs.
- Every task uses the same large model.
- Agents take minutes on simple tasks.
- Nobody has tested a new model this quarter.
- 01Paying twice for the same reading
Your AI rereads the same instructions and documents on every request. We store what repeats, so you pay for it once.
Caching - 02Too much on the desk
Each request carries pages the task never reads. We send only what the task needs. Less to read costs less, and often answers better.
Context engineering - 03A senior expert for every job
The largest model answers every question, even the simple ones. Each task gets the smallest model that does it well.
Model routing - 04Slow thinking for quick decisions
Sorting, tagging and reading a number from a document are quick decisions. System-1 models make them faster, and for less, than a large model.
System-1 models - 05Agents that go in circles
An agent calls the same tool ten times, or fetches a whole database to read one line. We read each step and remove the waste.
Tool-call inspection - 06Tools made for people, not agents
Many tools give an agent too much or too little. We rebuild them, so one call can replace several.
Tools for agents - 07Nobody knows what one answer costs
We show the cost of each answer, customer and feature. Budgets and alerts stop a surprise before it reaches the invoice.
Cost control - 08Last year's model
A cheaper model comes out almost every month. We test each one on your own tasks, and switch only when the quality holds.
Model evaluation
Public evidence
- Anthropic charges at least 90% less for input read from its cache. Anthropic: prompt caching
- Cloudflare moved one extraction step to a small model. That pipeline cost 49% less and ran 3.2 times faster. Cloudflare: Clairic
What could your AI bill be?
Four questions. The estimate shows before we ask for your email.
Your AI bill, before and after
An estimate. Change any answer.
1/4
Quality first. Then the bill.
We never cut cost by cutting quality. Each change must pass a test built from your real work.
- MeasureWe record what one answer costs today, and how good it is. Both sides sign the numbers.
- FindWe read the bill, the instructions and each step your AI takes. We rank the leaks by money.
- Fix and testEach change runs against the quality test first. If quality drops, the change does not ship.
- WatchBudgets, alerts and a monthly test of new models keep the bill down after we leave.
No change reaches your customers unless it passes the quality test.
What we need: read access to your AI usage, and one person from your team for 2 to 4 hours a week.
Three ways to work with us.
The same three ways as every Heinrichs.ai project. Here, the gain is the money your AI bill no longer costs.
Not sure which one? Check your AI bill: 4 questions →
It starts with the diagnostic week. It needs a saving large enough to share.
Paid from the gain
20% of what you save or earn in the first year. Then it is all yours.
We measure the gain in your own data, with a method we both sign.
You pay the 20% after we measure the saving, never more than the maximum in the proposal. Before that, you pay only the diagnostic week: if we measure no saving, the fixes cost you nothing. If the saving is less than 5 times the diagnostic week, the diagnostic becomes credit for other work.
What the diagnostic week gives you
- Your numbers today, and how we will measure them
- Every leak, scored and ranked
- A cautious estimate of the saving
- A proposal with the maximum you pay and the weeks of work
- Measured in your dataCost per answer, before and after, at the same quality score.
- Our gain guaranteeThe 20% follows the gain we measure, so a smaller gain costs you less. We also work up to 4 more weeks at no charge to close the gap. This holds if your team did its part: data and access on time, one person 2 to 4 hours a week, and the new process in use.
Diagnostic and solution in one week, for one fixed fee. The fee follows the gain.
- Before the weekWe agree the scope and the fixed fee.
- During the weekWe find the leaks and ship the fixes with your team.
- At deliveryThe fixes run on your real traffic. You keep the fixes and the week's report.
Work beyond the agreed scope needs your written approval. If the week finds a larger saving, it can become a project paid from the saving.
Open scope, in sprints of 1 or 2 weeks. You set the order.
- Before you startTell us what your AI does. We send the rate and an hour estimate.
- Each sprint1 or 2 weeks on the tasks you put first.
- After each sprintYou see the work and set the next order.
More terms
- Running costsYou keep paying your AI providers directly, at their prices. We take no share of your usage.
- What you keepYour license is permanent once the work is paid. If we stop operating, the code and your data pass to you.
- The next process, mapped at no chargeAfter the bill, the work itself: we map the processes your AI could take over, at no charge.
- If we disagree on a numberAn independent accountant, named in the contract, settles it with the signed method.
Your data and the evidence
- Your dataAn NDA and a data processing agreement (LGPD, GDPR) before we read any log. Read-only access. No AI training on your data. Access removed and copies deleted at the end.
- No lock-inAny provider, any model, any cloud. Your accounts stay yours, and so does every change we make.
How much can we save on our AI bill?
It depends on how your AI works today. The check above gives a cautious estimate in four questions. The diagnostic week measures it in your own usage.
Will the answers get worse?
No. First we build a quality test from your real work. A change that lowers the score does not ship, whatever it saves.
What is prompt caching?
Most requests repeat the same instructions and documents. The provider keeps that part ready and charges much less to read it again. Anthropic charges at least 90% less for cached input.
What is model routing?
A router sends each request to the cheapest model that does it well. Simple requests go to small models. Only the hard ones go to the largest model.
What is a System-1 model?
A small model that makes one quick decision in a single pass: a category, a number, or yes or no. It does not write text, so it is fast and cheap. Jev and Clairic are examples.
Do we have to change AI provider?
No. We work with OpenAI, Anthropic, Google, open models and any other provider. You keep your accounts and pay them directly.
What access do you need?
Read-only access to your AI usage and, for agents, their traces. An NDA and a data processing agreement come first. We never train AI on your data.
How long does it take?
One diagnostic week, then 1 to 4 weeks of fixes. The first fixes can go live in about 2 weeks.
Can our team do this alone?
Yes, with time. We bring a checklist of eight common leaks and build a quality test from your real work. Larger projects are paid only from the measured saving.
Send us one month of your AI bill.
Tell us what your AI does and what it costs. We reply with where it leaks and what the fixes cost.
Email us your AI bill Book a 20-minute call
From São Paulo, mostly remote, for Brazil and abroad. On request we visit during the diagnostic; travel outside São Paulo is billed at cost.
Who you talk to
Samuel HeinrichsFounder, São Paulo
Has built software since 2011 and led AI work since 2017. Postgraduate degree in Artificial Intelligence. Works in Portuguese and English.
We are taking our first clients and start at most 4 new projects a month. We reply within one business day, in Portuguese or English.