Bring us one job

Bring us one job.

The task that eats your team’s week. The capability your customers keep asking for. We’ll work out where AI fits, then build it.

Appoly’s AI division · Software delivery since 2017.

Tell us about your jobSee the work

No technical brief needed. Start with the work you want to make better.

A real call, recordedTechnical support line

Expertise on the end of the line.

Sam is the AI agent we built for a heating technical line. Here it helps a qualified engineer track down a boiler fault, live.

Caller
Sam, AI

The job: make technical guidance available while the engineer is on site. The qualified engineer remains responsible for the work.

Read the transcript
  1. Caller: There's a temperature drop across them, a differential for sure, so some heat is being transferred. I am noticing a bit of heat leakage down the central heating flow pipe.
  2. Sam (AI): Ah, that's a key detail. If you're getting heat bleeding into the CH flow during a hot water demand, that could point to the diverter valve, not fully sealing off

AI that clocks in.

Voice, documents, code, images, expert analysis and whole teams of agents. Each one has a defined job, for your team or inside the product your customers use.

  • It answers the phone.

    Takes calls from gas engineers on site, asks the right questions and talks them through the fault.

    A heating technical line · In their product

    EngineerI’m getting a resistance of about 7,000 ohms across it.

    Sam (AI)That’s quite a bit lower than expected for room temp. That would suggest the sensor’s reading hotter than it actually is…

    From a real call
    Hear the call

  • It reads the paperwork.

    Pulls every field out of supplier datasheets, whatever the layout, and flags the ones it isn’t sure about.

    A UK valve supplier · Operations

    Data sheetBV-340
    TypeBall valve100%
    Sub-typeTrunnion mounted100%
    Size8" × 6"100%

    954 fields in one run. 49 held for an engineer.

    One of 61 valves from a live run
    Watch the run

  • It fixes the bug it finds.

    Reads past tickets, live records and the source code, finds the cause and writes the patch.

    A global logistics platform · In their product

    Ticket Shipment stuck on “In transit” after delivery.

    - if (scan.type === 'DELIVERED')
  + if (isDeliveredScan(scan))

    Waiting for an engineer to approve

    Sample ticket. An engineer approves every fix.
    Follow a ticket

  • It inspects from a photo.

    Audits warehouse racking from a single photograph and marks the damage an inspector should see first.

    A warehouse racking specialist · Operations

    • Upright 3, level 1 Impact damage. Inspect before loading.
    • Beam locks All present.
    Sample image
    More on vision

  • It gives your customers an expert.

    Builds a full investment case for any property, then answers the customer’s follow-up questions in the product.

    A property investment platform · In their product

    CustomerIs this one better as a flip or a rental?

    Analyst (AI)I’ve compared both against local sales and rents. Here’s the case for each, side by side.

    1. Assess
    2. Build the case
    3. Answer
    Sample conversation
    See the capability

  • They work as a team.

    Six agents cover demand, operations, routes and customers, passing work between them. Your team decides which can act and which only suggest.

    A logistics operator · In build

    1. Demand Spots a jump in collections for Thursday.
    2. Operations Checks vehicles and crews against it.
    3. Routes Suggests a revised schedule.
    4. Your team Approves the change.
    Sample handoff
    Meet the team

Sixty-one valves.
954 fields in one run.

Document agent · A UK valve supplier

Engineers used to spend hours copying specifications out of supplier PDFs before they could quote. Now an agent reads the datasheets, fills in the fields and hands back only the ones it isn’t sure about.

About six minutes per datasheet, down from hours. 97.3% field accuracy across 120 test documents.

See all our work
Task 381
954 fields extracted. 49 held back for an engineer.

A live run, caught mid-job: the agent spots its own failed correction and retries it.

  1. Starting valve specification analysis
  2. Now that I have your approval, I will start the detailed extraction of all technical specifications from the documents.
  3. Detailed specifications extracted
  4. Now that I’ve processed your documents, I’ll double-check to see if any parts are missing their corresponding datasheets.
  5. Missing datasheet links resolved
  6. Validating your field correction
  7. Field correction validation failed: missing required field
  8. Scheduling retry #1
Parts
61
Fields
954
Need review
49
#Data sheetTagTypeSub-typeSize
1 70%BV-210 No datasheet10-XV-101ABall valve–10" × 8"
2 70%BV-210 No datasheet10-XV-101BBall valve–10" × 8"
3 70%BV-210 No datasheet10-XV-104ABall valve–12" × 10"
4 70%BV-210 No datasheet10-XV-104BBall valve–12" × 10"
5 70%BV-340 No datasheet10-XV-112Ball valveTrunnion mounted8" × 6"
6 70%BV-340 No datasheet10-XV-230Ball valveTrunnion mounted10" × 8"
7 70%BV-210 No datasheet10-ESDV-305Ball valve–10" × 8"
and 54 more parts
Task 381, replayed from the live system’s log. Amber marks what the agent flagged for a person to check. Datasheet codes and tags have been replaced to protect the client.

Real people.
Serious delivery.

Our team in Leamington Spa has shipped production software since 2017. We learn the job, test on your real work, connect to your systems and stay to support it.

You decide what each agent can touch and what needs a person’s sign-off. We plan with you how each role changes before anything goes live.

Meet the team and see how we work

James Merrix

Founder

Guy Stanton

Co-founder and Executive Chair

Mark Darling

Managing Director

John Wedgbury

Chief Technology Officer

Gurd Golar

Business Analyst

Alex Newberry

Business Development Manager