Self-hosted

Your cloud. Your data. Your AI platform.

SentientOne deploys single-tenant into your own AWS, Azure, GCP or on-prem environment. Agents, knowledge, MCP tools and the team workspace all run behind your perimeter — and our engineers do the deployment, the migration and every update after it.

Deployment · Migration · Training · Updates — all included

Deploys where your data already lives

  • AWS
  • Azure
  • Google Cloud
  • Private cloud
  • On-prem

Why teams self-host

Your staff already use AI. Just not on your terms.

Internal documents end up in public chatbots because it helps people finish the work in front of them. A sanctioned platform inside your own network gives them the same help on data that never leaves your control — and gives you the admin trail to show for it.

What you get

A complete AI stack inside your walls. Managed by us.

The same platform product teams run on sentientone.ai — only single-tenant, on your infrastructure, under your own security and compliance controls.

Runs inside your perimeter

The whole platform deploys into your environment. Prompts, uploaded documents and traces sit on infrastructure you already own and audit.

  • AWS, Azure, GCP, private cloud or on-prem servers
  • Full data residency — no processor you have not approved
  • Bring your own model keys and your own data stores
  • Built for finance, health, government and legal teams

One sanctioned AI for the company

A private workspace grounded in your own material, so staff stop pasting internal documents into public chatbots to get their work done.

  • Team chat with agents that read your docs and policies
  • Per-department agents, each with its own model and behaviour
  • Admin control over agents, keys, members and usage
  • Shadow AI loses its reason to exist

Agents that know your business

Point an agent at your documents, wikis and FAQs and retrieval does the work. No retraining, no fine-tuning bill.

  • Knowledge bases from internal documents, wikis and FAQs
  • Agents for HR, IT support, policy lookups and onboarding
  • Every agent isolated to your organisation

Your internal systems, connected

Expose the systems you already run through MCP and agents work out the rest — which tools exist, and when one is worth calling.

  • Plug agents into internal APIs, databases and tools
  • Staff pull a record or a report by asking for it
  • New tools appear without a schema to hand-write

We do the deployment, not your IT team

Standing it up inside your environment is our job, including moving across whatever you are running today.

  • Deployment, networking and configuration by our engineers
  • Migration of existing agents, data and settings
  • A planned go-live with no downtime
  • One named lead from kickoff to launch

Support comes with the licence

You own the deployment. You never run it on your own.

  • A direct line to our engineers, not a ticket queue
  • SLA-backed response times
  • Admin and team training included
  • Ongoing guidance on scaling and configuration

Self-hosted without version drift

Releases are patched into your instance on a managed schedule, so private never means two years behind.

  • New features reach your instance as they ship
  • Security patches and fixes on a managed schedule
  • No manual upgrades to plan around
  • Tested releases, so production stays quiet

Optional: make it yours

Match the platform to your internal brand if that matters to adoption.

  • Your logo, colours and internal domain
  • Branded login and workspace for staff

The whole platform, nothing held back

Custom agents, knowledge bases, MCP tool connections, the team workspace, the embeddable chat widget, the REST API, and full request tracing across auth, retrieval, tools, latency, tokens and cost.

How a rollout runs

Three steps, and almost none of them yours.

01

Scope the environment

One session on your cloud, network, identity provider and data rules. You get the deployment shape and what it needs from you.

02

We deploy and migrate

Our engineers install the platform, wire it to your identity provider, and move existing agents, data and settings across. Go-live is planned, not improvised.

03

Train, support, keep patching

Admin and team training, SLA-backed support from the engineers who built it, and new releases patched into your instance on a managed schedule.

Sign-in stays yours

Staff log in the way they already do. You keep users, groups, MFA and offboarding exactly where they are, and we map your SSO roles onto admin, team and member permissions.

  • OAuth
  • OIDC
  • SAML
  • Azure AD / Entra ID
  • AWS Cognito
  • Okta
  • Auth0
  • Google Workspace

What teams say

Every time you want to add AI to your app, you end up writing a ton of code — managing API keys, handling conversation history, engineering prompts, counting tokens. And then six months later a better model comes out and you have to refactor everything. SentientOne just removes all of that.
Daniel Marsh · CTO

FAQ

Before you involve procurement.

Something specific to your environment? Ask our engineers — they answer, not a form.

Can SentientOne run entirely on our own servers?

Yes. The full platform deploys into your cloud, private cloud or on-prem environment, and there is no runtime dependency on SentientOne-managed infrastructure.

Can staff sign in with our existing identity provider?

Yes. The platform plugs into the identity provider you already run — OAuth, OIDC, SAML, Azure AD / Entra ID, AWS Cognito, Okta, Auth0, Google Workspace or anything else. Staff sign in the way they already do, and you keep control of users, groups, MFA and lifecycle. We map your SSO roles onto admin, team and member permissions inside SentientOne.

Does any data leave our network?

The platform, your documents and your traces stay inside your environment. With bring-your-own-keys, prompts route from your network directly to the model providers you have chosen — on your account, at their rates, with nothing passing through us.

Do we still get updates if we self-host?

Yes. New features and fixes are patched into your environment on a managed release schedule, so there are no manual upgrades and no version drift.

Who sets it up and migrates our data?

We do. Our engineers handle deployment, networking, configuration and the migration of existing agents, data and settings, end to end.

What support is included?

SLA-backed support and a direct line to our engineering team come with the licence — no generic ticket queue. Admin and team training is included, along with ongoing guidance on scaling and configuration.

Get started

An AI platform your company actually owns.

Tell us your cloud, your identity provider and your data rules. We will come back with the deployment shape, the timeline and what it needs from your team.

  • Single-tenant
  • Bring your own keys
  • Migration included
  • SLA-backed support

You talk to the engineers who do the deployment.