Skip to content

Services

Four builds we price before we start.

Each one solves one problem, at a fixed price, in one to three 72-hour loops. Every one reuses something Axiom already built and runs in EquipTalkIQ.

See pricing
Axiom Routing / ServicesR-01 to R-04

01The four builds

One problem each. Scoped in writing.

The price is fixed and set in your statement of work after the intro call, based on your sources and scope.

R-01

Custom report or dashboard

The report your team rebuilds by hand, generated from the data you already have.

  • Up to two data sources
  • Every number with its calculation written down
  • Optional AI summary that links each sentence to its records
1 loopFixed price, set in your SOWWhat's included

Built beforeEquipTalkIQ's findings: each one generated by a tested rule, with the records that triggered it.

R-02

One integration

One system's records flowing into another system, or a database you own, on a schedule.

  • Written field map, tested on sample records
  • A log of every run and an alert when one fails
  • Safe to rerun: running it twice never creates duplicates
1–2 loopsFixed price, set in your SOWWhat's included

Built beforeEquipTalkIQ's ServiceTitan and ServiceTrade mappers, tested on sample data shaped from their API docs.

R-03

Single-workflow AI agent

One repeated task (triage, a draft reply, a document check, a follow-up) done by an agent that drafts.

  • A person approves every write and every outgoing message
  • Tested on at least 20 of your past cases before it goes live
  • Every AI call logged with its cost, under a monthly cap
2 loopsFixed price, set in your SOWWhat's included

Built beforeEquipTalkIQ's background agents, which run only after a person approves them.

R-04

Document intelligence

A set of PDFs (specs, manuals, contracts, closeout documents) turned into records you can search.

  • Answers that name the document and page
  • Fields extracted into a table or your system
  • Accuracy checked on a sample you pick
2–3 loopsFixed price, set in your SOWWhat's included

Built beforeEquipTalkIQ reads specifications, contracts and drawings and cites the document and page.

02How a build runs

The 72-hour loop.

Hours 0–24 · Scope

Write it down.

We write down with you what we'll build, which systems it touches and how we'll both know it works.

Hours 24–60 · Build

Build it, show you.

We build and test against sample data, and you see a working preview.

Hours 60–72 · Handoff

Hand it over.

You get the working build, the code made for you and a short note on how to run it. We review it together.

After handoff · option 1

Stop.

You keep the build. Defects against the agreed acceptance criteria are fixed free for 30 days.

Option 2

Another build.

A new statement of work for the next scoped problem.

Option 3

A monthly retainer.

Ongoing changes and new work, one request at a time. Pricing

03Systems we connect

Anything with a documented API or a reliable export.

The ServiceTitan and ServiceTrade mappers were built from the vendors' API documentation and tested on sample data. Every client connection is scoped in writing before we build it.

Field service

Jobs, appointments, assets, deficiencies and quotes.

ServiceTitan and ServiceTrade mappers built

Mail and files

Each user's mail and files, kept private to that user.

Google Workspace and Microsoft 365

Documents

Specs, manuals, contracts and closeout documents. Scanned pages need OCR, priced in the SOW.

Answers cited by page

Accounting

Invoices, jobs and costs from the system you already run.

By documented API or export

Spreadsheets and databases

The exports and tables your team keeps today.

By export or direct connection

Your system

If yours isn't listed, send us its API documentation and we'll tell you on the intro call whether it fits.

Scoped on the intro call

04How we build

The same rules on every build.

01

Answers with receipts.

Every AI answer or summary shows the records or pages it came from.

02

A person approves.

Agents draft. Nothing is written into your systems or sent outside your company until someone on your team approves it.

03

Costs you can see.

Every AI model call is logged with its tokens and cost, against a monthly cap you set.

04

Plain, proven tools.

Postgres with row-level security to keep each customer's data apart, a small web runtime, and direct calls to documented APIs.

05Security & confidentiality

Careful with what you share.

01

Mutual NDA.

Signed before you share anything non-public.

02

Least access.

Read-only or test-account access wherever the job allows. Credentials are stored encrypted and returned or revoked at the end.

03

Your data stays yours.

Used only for your build. We don't train AI models on it, and we use AI providers whose terms don't allow training on it either.

04

Returned or deleted.

At the end of the engagement we return or delete your data and confirm it in writing.

Start with one problem

Have a problem shaped like one of these? You'll know on the call.

A short intro call. If it fits a build, you'll have a written scope and a fixed price within a day of the call.

hello@axiomrouting.com