Skip to content

Deployment 02 · Maritime and shipping

Making complex maritime work conversational

A maritime organisation needed a simpler way for employees to work across a complicated internal process involving documents, systems, rules, permissions, and several steps.

Industry
Maritime and shipping
Ahum Labs’ role
Workflow research, conversational interface, integration, deployment
Status
Deployed
System diagram — natural-language interfaceDeployed
Employee · Plain language

Describes what needs to be done, in the words they would use with a colleague.

Behind the conversation
  1. Understand the intentWhat is actually being asked
  2. Retrieve relevant informationFrom approved sources
  3. Check permissionsMay this person do this?
  4. Complete the required stepsAcross several systems
  5. Return the resultIn the same conversation
The employee sees a simple conversation. The engineering remains behind it.
System mapPeople → Outcome
People
  • Shore staff
  • Operations
  • Approvers
Workflow
  • Multi-step process
  • Document handling
Systems
  • Internal software
  • Document stores
  • Permissions
Intelligence
  • Intent understanding
  • Retrieval
  • Step execution
Business outcome
  • Fewer menus, forms, and training days
People, workflow, systems, intelligence, and the outcome created.
01

The organisation

A maritime organisation working across vessels, documents, regulations, and internal software developed over many years.

02

The problem as presented

Employees needed a simpler way to work across a complicated internal process involving documents, systems, rules, permissions, and several steps.

03

What we found

The complexity was real and could not be removed. It sat across several systems and could not be redesigned without disturbing work that had to continue.

A traditional application would have introduced more menus, more forms, and more training—adding a layer on top of the complexity rather than hiding it.

04

The decision and trade-offs

A conventional interface would have been more predictable to build and easier to test. It would also have required employees to learn the architecture of the process before they could use it.

Employees already knew how to send a message and explain what they needed. We placed the workflow behind a natural-language interface and accepted the additional engineering required to interpret intent reliably.

05

The system

Employees could describe what they wanted in simple words. The system understood the intent, retrieved relevant information, checked permissions, completed the required steps, and returned the result.

06

How people used it

Employees did not need to learn the architecture. They only needed to explain what had to be done.

The employee sees a simple conversation. The engineering remains behind it.

07

Control and permissions

Permissions are checked before the system acts. A request that an employee is not authorised to make does not become an action.

08

Result

Outcome pending client approval

The measured result has not yet been approved for publication by the organisation involved. We will publish it here once it has been.

09

What we learned

Familiar behaviour is an interface. Where employees already know how to ask, approve, assign, and respond, those habits are the shortest route to adoption.

Next deployment

When the correct solution used less AI

Read the case study

Do not begin with a software requirement.Begin with the work that is not working.

Tell us where people lose time, information becomes difficult to find, decisions slow down, employees repeat the same work, or an existing system holds the organisation back.

We will study the operation before deciding what should be built.

No technical specification required