# Maplith for operations and delivery

How the AI agent an operations team already uses sees the work that was handed out with no delivery after it — shown on fictional companies’ exports, next to a purchase order that arrived after the shelf was empty.

Operations runs on handovers: a job assigned, a delivery uploaded, a purchase order received. Each handover is a date in one system, and the one that did not happen is a date that is missing — which no system reports. Your agent can see it — if it knows how the work flows, where each step is recorded and whether it may act. Maplith gives it exactly that; your agent acts in your systems with your credentials.

Acme Commerce is a fictional web shop built for our demo: the company, its people, its customers, the device brands, the suppliers and the competing chains are all invented, and nothing in the data comes from any customer.

LinguaNord ApS is a fictional translation agency built as our first test company: the company, its people, its customers and its translators are all invented, and nothing in the data comes from any customer.

## Between the systems

Recurring problems between the systems the work passes through.

- **Work handed out, with no delivery.** The process in the map shows the jobs that stop after assignment: an assignment on record and no delivery after it, a few of them assigned months before the last day in the data. (LinguaNord ApS · raised by a scan of its exports.)
- **The order that arrives after the shelf is empty.** In the worked example a purchase order due on 10 July arrived on 30 July, and the product it would have refilled was sold out from 14 to 29 July. (Acme Commerce · in its exported data.)

## Workflows

One workflow from the translation agency’s map. Everything in it ships today.

### Jobs assigned, with no delivery

Work is handed out, and nothing in the exports says it was delivered.

**Without Maplith.** A job’s assignment and its delivery are two dates in the project system, and some assignments are agreed in chat and reach the system late or not at all. In the worked example some jobs have an assignment and no delivery, a few of them assigned months before the last day in the data. An agent without a map has to read every job, decide what counts as late with nothing to compare against, and ask around.

**With Maplith:**

1. **Processes** — `get_process` for the job returns its happy path from the dates in the exports — assigned, then delivery uploaded — with the median time between the steps, and the variant that stops after assignment. A job that has waited far longer than the median stands out against the business’s own pace.
2. **Route** — `route` from the job to its delivery says which system holds each and by which key; the map marks the job’s join to its delivery strong and its join to the assignment agreed in chat weak, so the agent knows which one to trust.
3. **The action gate** — `request_action` for a follow-up the business declared — for example, a message to the job’s project manager — decided by the key’s policy.
4. **Record and value** — `record_action` writes what was done, so the next task context on the same thing shows it.

Worked example: [LinguaNord ApS](/industries#linguanord).

## The pieces

| Piece | Tool | What it gives the agent | Status |
|---|---|---|---|
| Processes | `get_process` | How the work actually flows, from the dates in the exports: the happy path, the time between steps, and every variant that deviates. | live |
| Route | `route` | The shortest verified join path from a start object to a target: which system, which key, how much to trust each hop. | live |
| The action gate | `request_action` | Allowed, pending a person, or denied — before the agent acts, with a grant good once. | live |
| Record and value | `record_action` | What was done, so the next judging does not raise it as though nobody had looked — and counted in “What you got”. | live |

## The worked examples

- **[Acme Commerce](/industries#acme)**, a web shop selling phone and tablet accessories. Its situations are in twelve months of exported data, checked against the files by a script with no model involved. These pages say what is in the data, and publish no score for what a scan of it finds. [Maplith for e-commerce](/industries/e-commerce).
- **[LinguaNord ApS](/industries#linguanord)**, a translation agency. The findings and processes quoted from it come from a recorded scan of its exports. There is no public map or key for it.

Every department: [Solutions](/solutions). The worked examples: [Worked examples](/industries).