AI

An AI agent for technical support and 3D equipment visualisation

The context

The same equipment manufacturer now had the data from its fleet. That left a problem telemetry does not solve: interpreting what it says.

A technician on an intervention looked for answers in PDFs, in a car park, on an unreliable connection. The documentation exists, and it is good — but it is spread across manuals, procedures and a parts catalogue, each varying by version and by equipment configuration. Finding the right page assumes you already know which one you are looking for.

In the back-office the problem was the mirror image: the fleet-management and after-sales interfaces had grown to the point where a new joiner had to learn the menu tree before being able to answer a customer.

What we did

We built an assistance layer on top of the existing platform, grounded in the technical reference material rather than in a model’s general knowledge.

An agent grounded in indexed documentation and in live fleet dataThree sources feed the agent, reached two different ways. The technical documentation is ingested into Knowledge Bases, which chunks it, embeds it and stores the vectors in Amazon S3 Vectors: it changes rarely, so it is indexed ahead of time. The parts catalogue with its compatibility rules and the fleet telemetry holding the history of the specific machine are reached live through an MCP server, because a pre-computed index of them would be out of date. Both paths converge on the agent, which runs on Amazon Bedrock and carries the source of every answer. It is consumed by the technician on site for troubleshooting, inside the back-office as guided navigation, and by the support desk with the full equipment history to hand.Grounding the agent in the reference materialTechnical documentsmanuals, proceduresParts cataloguereferences, compatibilityFleet telemetrythis machine’s historyKnowledge Basesindexed on S3 VectorsMCP serverthe live sources, queried as they areAgent on Amazon Bedrock — answers cite their sourceTechnician on sitetroubleshootingBack-officeguided navigationSupport deskfull equipment history
Two retrieval paths, because the sources differ in one respect that decides the design: documentation is worth indexing ahead of time, machine state is not. The machine history is in the list at all because without it the agent answers about the model rather than the unit in front of the technician.
  • The company’s technical documents indexed with Knowledge Bases. Manuals, procedures and service notes are ingested from S3, chunked, embedded and made retrievable by the service, which then serves as the agent’s source. The vector store behind it is Amazon S3 Vectors, which avoids running and paying for a dedicated vector database at this scale. Using a managed knowledge base rather than assembling the ingestion, chunking and retrieval ourselves is the difference between a component the client can re-index by dropping a new revision in a bucket and a pipeline someone has to keep alive.
  • An MCP server over the sources that stay live: the parts catalogue with its compatibility rules, equipment configurations, and the fleet telemetry. These are queried as they are rather than indexed, because a pre-computed index of stock, compatibility or machine state is a confidently wrong answer waiting to happen. One interface over sources that did not share a shape, which also avoids re-integrating each of them into the next tool that wants to read them.
  • A troubleshooting agent on Amazon Bedrock, drawing on both: the indexed documentation for how the equipment is meant to work, and the history of the specific machine for how this one has actually been behaving. The distinction matters: an answer about the generic model is almost always useless, because configurations differ from one unit to the next.
  • Navigation assistance in the back-office. A user states what they need in plain language and lands on the right screen with the filters already applied, instead of learning where the function lives. The menu tree stays available for those who know it: the agent is a shortcut, not a replacement.
  • Generated 3D exploded views. From the assembly model matching the equipment’s configuration, the agent produces an exploded view of the sub-assembly concerned: which parts come out, in what order, and under which reference. A stock diagram would have been easier to serve, and wrong as soon as a configuration departs from the nominal case.
From the assembly model to an exploded view of the right sub-assemblyThe assembly model for the equipment configuration is the starting point. The agent locates the sub-assembly the symptom points to. An exploded view of that sub-assembly is generated on demand, showing which parts come out and in what order. The part reference comes with it, so it can be ordered before anyone travels.From the assembly model to the part in handAssembly modelper configurationSub-assemblylocated by the agentExploded viewgenerated on demandPart referenceordered before travel
The last step is the one that pays for the chain: an exploded view is interesting, a part reference ordered before the van leaves is what removes the second visit. Generating the view per configuration rather than picking a stock diagram is what makes the reference the right one.

The outcome

The technician arrives with the part. In this trade that is the only measure that really counts: the second visit, the one that costs a full trip because the right reference was not identified in time, is becoming rare.

The support desk answers faster, and a new joiner becomes useful in days rather than weeks, because they no longer have to memorise where things are before they can help anyone.

The most useful lesson is about trust. An agent that is confidently wrong costs more than an agent that declines to answer: it sends someone to site with the wrong part. So every answer cites the document and section it came from, and the agent says explicitly when the reference material does not support a conclusion. Knowledge Bases returns the passages it actually retrieved, which is what makes that citation worth something: it is a fact of the retrieval rather than something the model was asked to supply, and a model asked to cite its sources will invent one. That is not a legal precaution, it is what keeps a technician using the thing after its first mistake.

All case studies

Talk to us directly

Write to us directly. No salesperson, no qualification call. One of us two replies, within 24 hours.

contact@onescale.io

We reply within 24 hours, and it is one of us who replies.

Based in Lyon, working with clients in France, across Europe and internationally.