Case studies

What we build, step by step.

Each build, step by step: the problem, the workflow, the tools it connects and the checks that keep it accurate.

Book a free 30-minute call

Businesses

AI automation we build for service businesses.

Each one starts in tools a service business already pays for and adds only what's missing.

Website formGmailHubSpotSlack
  1. 1Enquiry arrives
  2. 2AI reads it
  3. 3CRM updated
  4. 4Owner assigned
  5. 5Reply drafted

AI automation · Sales inbox

4Tools connected

Every enquiry filed, assigned and answered in draft.

AI reads each enquiry from the form or the shared inbox, files it in the CRM, routes it to the right person and drafts a first reply for them to send.

How the enquiry-to-CRM automation works
Bank feedQuickBooksClickUpGmail
  1. 1Payment lands
  2. 2Matched to invoices
  3. 3Clear matches closed
  4. 4Unclear ones queued
  5. 5Reminders drafted

AI automation · Finance

4Tools connected

Payments matched to invoices, with only the odd ones left for a person.

Bank payments are matched to open invoices as they clear. Clear matches close themselves, and the odd ones go to a short review list with the AI's best guess.

How the payment matching automation works
HubSpotAsanaGoogle DriveQuickBooksSlack
  1. 1Deal marked won
  2. 2Project set up
  3. 3Files and access
  4. 4Billing ready
  5. 5Team told

AI automation · Operations

5Tools set up

One deal marked won, and five tools are ready for the client.

When a deal is marked won, the project, the shared folder, the first invoice and the team channel are all set up from the deal record, with nothing retyped.

How the client onboarding automation works

Startups

AI features we build for startups.

Each one is built in production code, with the parts that keep it working once real users arrive.

Next.jsPostgrespgvectorClaude APIIntercom
  1. 1Ticket arrives
  2. 2Sorted by topic
  3. 3Docs searched
  4. 4Reply drafted
  5. 5Agent sends it

AI app · Customer support

100%Answers sourced

Support answers from your own docs, or an honest "I don't know".

Tickets are sorted by topic and urgency, answered from your own help articles with the sources linked, and sent by a support agent who can edit the draft.

How the support answer assistant works
ReactNode.jsS3GeminiZod
  1. 1File uploaded
  2. 2Fields read
  3. 3Checked in code
  4. 4Unsure fields flagged
  5. 5Saved to the app

AI app · Document processing

EveryField validated

Upload a contract, get clean data back.

Users upload PDFs, scans or photos. The AI pulls out the dates, names, totals and terms into a fixed format, and anything it isn't sure of is highlighted for the user to confirm.

How the document extraction feature works
Next.jsPostgresOpenAI APIRedisPostHog
  1. 1User asks
  2. 2Permissions checked
  3. 3Answer built
  4. 4Costs capped
  5. 5Usage tracked

AI app · In-product assistant

EverySearch permission-checked

Answers from each user's own data, and nobody else's.

Users ask questions in plain English inside the product. Answers come only from records they're allowed to see, with links back to each one and a cap on AI costs.

How the in-app data assistant works

Working in 21 days from access, or your build fee back. Half paid only when it works. Full terms

Book a free 30-minute call