unthinkable
Case study · Infrastructure investor, London, UK · 17 international offices

The investor's employees ask questions and get answers inside a secure company platform, with the context of their own role.

The investor wanted its people across 17 offices to find information quickly and share knowledge between departments, without sensitive data leaving a secure environment.

London, United Kingdom

What changed

300Kquestions asked (as reported)
2000+active users (as reported)
100Kthreads shared (as reported)
What stood in the way

What stood in the way.

What stood in the way
  • The investor wanted generative AI for its employees, but only on a platform that is secure and open to authorised staff alone.
  • It needed a partner for the strategy, architecture and design as well as the build.
What changed
  • 300Kquestions asked (as reported)
  • 2000+active users (as reported)
  • 100Kthreads shared (as reported)
What we built

What we built.

  • An employee web portal to ask questions, giving background and role so answers fit the person asking, on Microsoft AI APIs in Azure.
  • Threads to return to earlier conversations, search across them and delete chat history.
  • An admin portal with user logins, a dashboard of users, active users, questions and last interactions, and daily and monthly reports with sorting and filters.
  • React, Node.js and Azure Cosmos DB.
  • Microsoft Azure
  • Microsoft AI APIs
  • React
  • Node.js
  • Azure Cosmos DB
What it shows

What this work shows about us.

AI kept inside the company

Only authorised employees reach it, in the client's own cloud, with admins seeing how it is used.

Answers that fit the person

Each question carries the asker's role and background, so answers suit what they actually do.

Want to build a similar secure internal generative AI platform? Tell us what you want to change. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with the right team.

Start your plan ↗
If we built it today

If we built it today, with AI.

What we would add now, moment by moment.

Finding an answer

An agent would answer from the firm's own approved documents and show the source for each point, and say plainly when it does not know. People still decide what to act on.

Doing the next step

An agent would go beyond answering: draft the memo, fill the request or file the note the employee asked about, for them to check and send. Access rules stay as they are.

Ask Unthinkable

Want this for your business?

Tell us what you run or what you're building. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with a senior AI engineer.

Send us a note instead

Required

Thanks. A senior AI engineer will reply to you.
u
Ask UnthinkableClarity first, then the right team
Your conversation

An AI agent. This prototype uses prepared answers and sends nothing.