Our work

Work you can examine.

A worked example, reference architecture and exploratory concepts. The status of each is stated so you can judge what the material actually demonstrates.

A diagram separating answer claims with source support from claims without it
Check the support for each claim · Concept diagramThe sample source requires review every 12 months. It supports an annual-review statement, but does not establish a requirement for finance approval.

Start with the method

Inspect an answer review.

Our worked example shows an invented policy, questions, deliberately authored answers and the reasons for each judgment. The worksheet is available to download.

It demonstrates a review method. It is not a client case, a model benchmark or evidence of a deployed system.

Open the worked example ↗
A diagram separating answer claims with source support from claims without it
Check the support for each claim · Concept diagram
Reference architecture

Private RAG

Exploring how source selection and answer generation should be examined separately.

Explore the reference work ↗
Illustrative laboratory equipment beside a procedure and incident-review workstation
Laboratory incident review · AI-generated setting, not an actual facility
Exploratory concept

Sentinel

Investigating how evidence and procedure context could support a laboratory incident reviewer.

Read the research question ↗
A diagram separating answer claims with source support from claims without it
Check the support for each claim · Concept diagram
Proposed concept

EvidenceDesk

An interface concept for examining answer claims beside their supporting material.

Read the concept ↗

ClinicFlow is an archived proof of concept. See the archive note.

Questions still being shaped.

Our quantum computing direction is at exploration and proposal stage. Scheduling and resource allocation are possible starting points; no project has been selected or run, and no funding award or validated result is claimed.

When measured work is published, its task, baseline, sample and limitations should be visible. An attractive interface alone is not evidence that a method is useful.

For a question in your own organisation, explore AI Research & Feasibility, AI Proof of Concept or Model Evaluation & Benchmarking.

A considered next step

What do you need to find out?
Start with the question.

Tell us the task, the material available and the decision you need to make. We will discuss whether a focused investigation is the right next step.

Discuss your question hi@manaslabs.co.uk
01   Describe the task02   Agree the investigation03   Review the evidence