30 minutes with James Rooney to discuss your team, workflow and first scope.
For product, operations and technology teams exploring an assistant for internal knowledge, specialist guidance or finding information across an agreed set of company sources.
Make the supporting passage and document easy to open. Colleagues can check the context and apply their judgement to the decision in front of them.
Design retrieval around identity and document permissions. Test access boundaries alongside answer quality so the assistant fits the way your organisation shares information.
Agree a set of questions with the people who know the material. Check whether the assistant finds relevant evidence, answers appropriately and recognises gaps.
Use a 30-minute call to discuss the outcome your team needs.
Talk it throughStart with a recognisable task and a manageable source collection. Identify the content owners, the users and the information each needs to do their work.
Our approach uses retrieval-augmented generation, or RAG, where useful: find relevant material, then use it to support the response. Test missing, outdated and conflicting information too.
Plan how sources get updated, questions get reviewed and changes get tested. Keep the evaluation examples and ownership clear as the assistant develops.
Ask how this approach would fit your data, systems and team on a 30-minute call.
Talk it throughTenhaw's working proof of concept at Placing Platform Limited extracted and enriched PDF information for business intelligence. A client engineer paired throughout the two-week build, connecting the implementation to the people responsible for the data.
Read the document-to-BI proof of conceptThe guide explains how we would design source access, permissions and evaluation for an enterprise knowledge assistant, including the questions to settle before choosing the technology.
Explore our approach to enterprise retrievalBook 30 minutes to ask James about the work and how it relates to your situation.
Talk it throughAgree the workflow, deliverables and team before work starts. These are the existing engagements this use case can sit within; prices exclude VAT.
Fixed price · £44,000
4 weeks
Find the AI workflows worth funding. Test them with your team.
See scope, team and termsFixed price · £20k–£55k
2–4 weeks
Build and test one AI workflow in your own tenancy.
See scope, team and terms£70k–£85k / month
6–12 months
Build enterprise AI with an embedded team your people learn from.
See scope, team and termsUse a 30-minute call to decide which scope is a sensible starting point.
Talk it throughA 30-minute discovery call with James Rooney. Bring a workflow, product or team challenge and we will discuss a sensible first scope.
most start with a fixed-price AI Readiness Audit · £44,000 · 4 weeks · working prototypes
No time that suits? Email hello@tenhaw.com. Calendar not loading? Open it on cal.com.
It should surface the conflict and the relevant sources for review. The content owner decides which policy applies. Version, publication date and ownership help the assistant recognise differences, while evaluation examples check that it escalates the conflict appropriately.
Yes. We can evaluate retrieval and responses using a set of real questions before investing in the interface. That gives the team evidence about the sources and answers, which then informs where the assistant belongs in their working day.
Choose questions connected to a defined task, with source material and an owner who can assess the answer. Include frequent requests, difficult exceptions and questions that should be referred to a person. That becomes the initial scope and evaluation set.