Skip to main content
VectisFlow
VectisFlow

Services

Four services, one forked pipeline.

We are engaged to build and hand over. Each service is a different entry point into the same pipeline, and most engagements touch two of them.

What we are engaged to do

Pick the entry point; the pipeline is the same.

Engineering and extraction usually run together. Strategy comes first when nobody has opened the estate yet. Assurance runs alongside everything and is the one most often left out.

  • Service 01

    Data & platform engineering

    Warehouse and lakehouse build, migration, ingestion, modelling, and the contracts that keep the numbers trustworthy enough to act on.

    See how it works

  • Service 02

    Intelligence extraction

    Retrieval, extraction and enrichment over the documents, tickets, contracts and transcripts that never reached the warehouse.

    See how it works

  • Service 03

    Data & AI strategy

    A sequenced plan grounded in what your data can actually support, with the disqualifying constraints found before the budget is committed.

    See how it works

  • Service 04

    Evaluation & assurance

    Evaluation harnesses, adversarial testing and data contracts, so both the answers and the numbers underneath them can be checked.

    See how it works
An aisle between rows of server cabinets in a data centre, cabling visible along one side.

The pipeline

Where each service sits.

Platform engineering builds the spine both branches run on. Extraction takes the retrieval branch, strategy decides which branch earns the next quarter, and assurance sits on the join because both branches have to be checkable.

The VectisFlow pipeline: one shared spine, forking into two branches that rejoinSurvey, Ingest, Parse, Model, Quality run once and are shared. The path then forks: Semantic, Publish on the analytics branch, Chunk, Enrich, Index, Retrieve, Ground on the retrieval branch, both running in parallel. The two rejoin at Evaluate, Review. Access control resolves from the index into retrieval on every query. Evaluation feeds back into chunking and into the model the analytics are built on, and human review feeds back into evaluation.01Survey02Ingest03Parse04Model05Quality06Semantic07Publish08Chunk09Enrich10Index11Retrieve12Ground13Evaluate14Reviewthe same labelled set tunes the modelACLs resolved per query, before rankinglabelled set tunes chunking and retrieval
Solid: the forward path, shared then forked. Dashed: the three loops that decide whether it improves.

Engagement shape

How engagements are shaped

We are engaged to build and hand over, not to stay. Every engagement is scoped so that the people who will run the system afterwards have changed it themselves before we leave.

  1. 01

    Discovery, 2 to 3 weeks

    A survey of the real sources, structured and documentary alike, with permission modelling and an evaluability screen. Ends with a written recommendation, including the recommendation not to proceed where that is the honest answer.

  2. 02

    Build, 8 to 16 weeks

    Two-week increments against a working system, not a sequence of documents. Your engineers are in the repository from the first increment; a handover that begins at the end is a handover that fails.

  3. 03

    Assurance, in parallel

    The harness is built alongside the pipeline, not after it, whether it measures answers or the numbers behind them. Nothing goes to users without a measured baseline it can be compared against.

  4. 04

    Handover and support

    Runbooks, a rollback path that has been executed at least once against real data, and a defined support window that ends. If you need us permanently, something has gone wrong with the handover.

What the first 90 days looks like

Start here

Not sure which of these you need?

Most people are not, at the point they get in touch. Describe the estate and what you want out of it and we will tell you which of the four this actually is, including when the honest answer is none of them yet.

Response time
One working day, from a person who has read it.