01 · Knowledge
What your organization knows, available to everyone
Your documents, methods, and past work become a knowledge base that anyone on the team can query in natural language, with citations to the sources used.
Private AI as a service
Your own AI, powered by your organization’s knowledge
Like ChatGPT or Claude, but yours: connected to your organization’s knowledge and methods, running on infrastructure you control, with answers that cite their sources. We implement, customize, host, and maintain it.
Astraed in action
A different AI for every process
See how a team switches between specialized copilots and gets an answer grounded in its own methodology.
“Since we started using Astraed, everything changed: how we work, our speed, and our quality. Today we cannot imagine operating without it.”
What changes
Your organization’s knowledge becomes a shared capability
02 · Control
Your data, in your environment
The platform runs on infrastructure you control. Connections to external services follow your policies, and each person sees only what they are authorized to access.
03 · Operations
No in-house technical team required
You do not need to hire engineers or manage servers. We install, configure, back up, and keep everything running.
The service
Your private AI workspace, live in four weeks
Astraed installs, personalizes, launches, and operates the complete workspace. Your team provides the knowledge and workflows; we handle the technical work.
Before we begin
One project owner · OpenRouter payment method · Initial knowledge base · Workflow descriptions · Access to approved services
About five client hours
-
Week one
Architecture and installation
- One-hour kickoff
- Install the Astraed workspace
- Configure models, security, and backups
- Approve the plugin blueprint
ResultCore workspace online and scope approved
-
Week two
Knowledge and personalization
- Structure the initial knowledge base
- Build one plugin containing four custom agents
- Develop each agent’s skills
- Connect Notion, databases, and approved services
ResultFirst working version ready for testing
-
Week three
Pilot and refinement
- Run real client workflows
- Validate answers and citations
- Test permissions and connections
- Refine agents from feedback
ResultClient-approved system ready to launch
-
Week four
Launch and operation
- Add users and final permissions
- Train the initial team
- Verify monitoring and backups
- Open ongoing support
ResultFully operational Astraed workspace
From week five
Managed as part of the service: maintenance, backups, platform and plugin updates, support, knowledge updates, and up to four new agents with their skills each month when required.
Four weeks covers the standard scope. During kickoff, we agree on the success checklist. If the standard implementation does not meet it by day 30, we continue working at no extra cost. Larger knowledge bases or additional plugins and agents may require a longer timeline agreed during discovery.
Solutions
Adapted to how your sector works
The same foundation, adapted to what your clients, regulators, or authorities require.
- Consulting firms Methodologies, proposals, and past projects available to the entire team.
- Law firms and legal teams Firm precedents and playbooks, with matter-level permissions.
- Enterprises Approved AI and permissioned internal knowledge for every function.
- Universities and research Teaching, research, and administration with group-based access.
- Government & defense On-premise or disconnected operation inside your perimeter.
Sovereignty and security
Your data, your models, and your keys
You decide where your data lives, which models can reach it, and who has access.
- Infrastructure under your control
- Your private cloud, your servers, or a disconnected environment. You decide where your data lives and who has access.
- Permissions based on the work
- Access is managed by user, group, and project. Each team sees only the models, documents, and copilots it is authorized to use.
- Models your organization approves
- Approved external providers or open models running locally, in the combination your organization defines.
The comparison
Why not just use ChatGPT?
Where your data lives
In their cloud, under their terms.
On your infrastructure, under your terms.
Cited sources
Sometimes, from the web.
Always — from your documents and, with PathMBA, a curated business library.
Your methodology
Only if someone structures it.
Loaded during implementation.
Who keeps it working
Your IT team.
We do, in four weeks.
The honest answer: if you only need a chatbot for your files, a generic enterprise plan is cheaper. If your deliverables need verifiable sources and your clients require real confidentiality, you need AI of your own.
Investment
Clear pricing, no hidden usage
You pay to implement and operate Astraed. Model usage and infrastructure remain visible and under your control.
Initial implementation
Four-week deployment, infrastructure configuration, knowledge loading, permissions, copilots, and pilot.
US$5,000
one-time payment
one-time payment
Monthly operation
Hosting, monitoring, backups, updates, support, and continuous refinement. Base plan for 10–20 users; pricing adjusts by volume for 21–40; for more than 40, we prepare a proposal.
From US$1,500/month
based on users
based on users
Model tokens
Your organization pays for its tokens directly to the provider or through Astraed at cost.
Based on usage
no additional markup
no additional markup
An expert business knowledge source connected to your organization’s private AI. Included at no cost with an annual Astraed contract.
US$599/month
included with annual contract
included with annual contract
For 15 users on an annual contract, the cost is about US$100 per user per month with PathMBA included — less than one junior analyst, working for your entire organization.
For multiple business units, disconnected environments, or special security and support requirements, we prepare a custom proposal.
Calculate your caseQuestions
Private AI, explained
What exactly am I buying?
A working capability, not a license. Astraed implements, customizes, hosts, and maintains your organization’s private AI. The stack combines proven open-source software with the models you approve and the infrastructure you choose.
How is this different from giving my team ChatGPT?
It knows your organization, your data stays on infrastructure you control, and access is managed by team and project with usage records.
Do I need a technical team?
No. Installation, monitoring, backups, and updates are part of the service. If you have an IT team, we work within its policies; if you do not, one is not required.
Where does our data stay?
In your deployment infrastructure: your private cloud, your servers, or a disconnected environment. Data reaches an external service only when your organization configures it.
How long does it take, and what do you need from us?
The initial implementation plan is designed to reach launch in four weeks. From your side: access to the chosen infrastructure, documents that define how your organization works, and a few hours from your team for the pilot.
How much does it cost?
Implementation costs US$5,000 once. Monthly operation starts at US$1,500 for 10–20 users; pricing adjusts by volume for 21–40, and we prepare a proposal for more than 40. PathMBA is included at no cost with an annual contract. Your organization pays for its tokens directly to the provider or through Astraed at cost.
Next step
Evaluate your case in 30 minutes
We review your operation, the knowledge you have documented, and estimate the case using your numbers.
Evaluate your case in 30 minutes