Self-Hosted · On-Premise

The AI chatbot thatnever leaves home.

The same Outlearn agent that reads your knowledge, takes real actions, and handles every customer — running entirely on your infrastructure, behind your firewall, under your control.

Data never leavesYour own modelsFull source access

Trusted by teams at

Amazon logo
TCL logo
Ericsson logo
Upstart logo
Appcues logo
ZoomInfo logo
Shipt logo
Virgin Mobile logo

Your data, your walls

Your data never leaves the building.

With self-hosted Outlearn, the entire platform runs inside your environment. Every conversation, document, embedding, and model call stays behind your firewall. There is no copy of your data on our servers — because your data never reaches them.

  • Nothing is sent to Outlearn — not telemetry, not content
  • Deploy in your own cloud, private VPC, or fully air-gapped
  • Meet GDPR, HIPAA, and data-residency requirements with ease
  • Your security team holds every key and reads every log

Inside your boundary

Your infrastructure

Firewall

Outlearn app

Ruby on Rails

Your database

Conversations & docs

Embeddings

Private vector store

Your models

On your GPUs or endpoint

Anthropic Icon Streamline Icon: https://streamlinehq.comOpen Ai Logo Light Streamline Icon: https://streamlinehq.com

Open-weight or your own private endpoint — your call

Conversations, documents, embeddings, and model calls never cross this line.
Outlearn Cloud — receives nothing

Complete control

You own everything.

Not a tenant. Not a seat. Your deployment, your models, your code — answerable to no one but you.

Your data never leaves

Conversations, documents, and embeddings live entirely inside your network. Nothing is sent to Outlearn — there is no copy on our side to leak, subpoena, or breach.

Bring your own models

Run open-weight models on your own GPUs, or point Outlearn at your private Azure OpenAI or Bedrock endpoint. Your prompts go only where you decide.

The code is yours

You get the full source — no black boxes, no obfuscation. Read it, audit it, run it in air-gapped environments. It's your deployment to inspect, line by line.

Complete transparency

Every request, retrieval, and model call is observable in your own logs. Your security team sees exactly what the agent does, with nothing hidden behind a vendor API.

Built on modern Ruby on Rails

A mature, battle-tested stack your engineers already understand — easy to deploy, monitor, and reason about. No exotic runtime, no surprises in production.

Compliance & data residency

Keep regulated data in-region and inside your own compliance boundary. Built for teams answering to GDPR, HIPAA, and strict procurement reviews.

Cloud or self-hosted

Same agent. Your choice of home.

Start in our cloud and switch later, or run self-hosted from day one. The product is identical — the only difference is whose servers it lives on.

Outlearn Cloud

Fully managed by us. The fastest way to get a working agent live — we handle hosting, scaling, and updates.

  • Live in minutes, nothing to provision
  • We run the infrastructure and updates
  • SOC 2 Type II, encrypted end to end
  • Great for most teams
Maximum control

Self-Hosted

Runs entirely inside your environment. For teams who need data, models, and code to stay fully under their control.

  • Your data never leaves your network
  • Bring your own models and endpoints
  • Full access to the source code
  • Built for compliance and data residency

What it does

The full Outlearn agent — hosted by you

Trained on your knowledge

Connect your help center, docs, and past tickets. Outlearn answers from your content — not generic guesses.

Takes real actions

Look up an order, open a ticket, or trigger a workflow. Your agent does the work, not just the talking.

On every channel, 24/7

A web widget, email, Slack, Teams, and more — one agent handling every customer, around the clock.

How it works

Runs entirely on your infrastructure.

Instead of running on Outlearn's cloud, the whole platform is deployed inside your environment. Our team works alongside yours to provision, configure, and keep your deployment current — you're never on your own.

  • Ships as containers that fit your existing stack
  • Deploy in your own cloud, VPC, or on-prem
  • Set up and maintained together with our team
  • Scales with your volume — no per-conversation caps

What you'll need

Your own environment

Run it in your cloud account, a private VPC, or your own data center.

A container runtime

Outlearn ships as containers, so it fits the infrastructure you already run.

Somewhere to store your data

A database and file storage for your knowledge and conversations.

Our team alongside yours

We help you provision, configure, and keep your deployment up to date.

Not sure what fits your setup? Our team scopes the exact requirements with you.

The same integrations, running on your infrastructure

Zendesk
Notion
Confluence
Help Scout
Shopify
Slack
SharePoint
GitHub
Google Calendar
Calendly
Box
Figma
Zendesk
Notion
Confluence
Help Scout
Shopify
Slack
SharePoint
GitHub
Google Calendar
Calendly
Box
Figma
Salesforce
Google Drive
Intercom
Gmail
Stripe
MS Teams
Dropbox
Outlook
Linear
Asana
OneDrive
Supabase
Freshworks
Google Chat
Salesforce
Google Drive
Intercom
Gmail
Stripe
MS Teams
Dropbox
Outlook
Linear
Asana
OneDrive
Supabase
Freshworks
Google Chat

Get started

Own your AI, end to end. It's yours.

Tell us about your environment and our team will scope a self-hosted rollout and a quote that fits — no data leaves your walls, ever.

Self-Hosted AI Chatbot — Run Outlearn on Your Own Infrastructure