Small company or consulting team
An initial internal deployment with occasional generation and exports.
- CPU
- 4vCPU
- Memory
- 8GB
- Working storage
- 40GB SSD
One application instance, a dedicated database, scheduled backups and a named administrator.
Your infrastructure. A considered deployment.
Self-hosting, with clear requirements.
Run the DocuEngine web application on infrastructure you manage. Plan around document generation, digital system engineering, PDF exports and your team’s real workload—not employee count alone.
Assisted deployment—not a desktop installer
A standalone source build has been prepared. Customer-specific authentication, domains, AI configuration and Arabic export fonts must be reviewed and validated before go-live. Self-hosting does not mean fully offline operation or automatic high availability.
The static website package is separate from the application build; uploading it does not install DocuEngine.
Minimum software and operational prerequisites
- 01
Server runtime
A 64-bit Linux server or VM capable of running Node.js 24 and Chromium. pnpm 9+ is required when installing or building from source; validate the chosen OS and browser libraries during setup.
- 02
Database
A dedicated PostgreSQL 16 or compatible database, with persistent storage, restricted access and a tested backup/restore procedure. Never reuse another customer’s or the live service’s database.
- 03
Document rendering
Headless Chromium with its system libraries and the supplied document fonts. Verify Arabic shaping, RTL layout and PDF/DOCX exports on the target machine before acceptance.
- 04
Access and network
An HTTPS domain, correct application/API routing, approved identity configuration and controlled outbound access to enabled AI and other external services. Users access the application through a current web browser.
- 05
Optional connected services
Outbound email, error reporting and shared rate-limit/delivery state need their corresponding services configured. The current Redis integration uses an Upstash-compatible HTTP endpoint; a plain Redis TCP server is not a drop-in replacement.
- 06
Operations
A named technical administrator, secure configuration, restart supervision, log review, disk monitoring, off-host backups and a planned maintenance window.
Technologies behind the self-hosted application
This is the application stack in the prepared standalone build. Identity, AI, email and optional shared-state services need deployment-specific configuration.
- 01 / STACK
Web client
React and TypeScript, routed with Wouter, built with Vite and styled with Tailwind CSS. TanStack Query manages API data; Framer Motion animates the interface; Lucide React supplies icons; React Flow renders interactive process maps; Clerk provides sign-in.
- 02 / STACK
Application API
Node.js 24 and Express 5 run the TypeScript REST API; Zod validates structured request data.
- 03 / STACK
Database and data access
PostgreSQL 16 or compatible, accessed with node-postgres (pg) and Drizzle ORM.
- 04 / STACK
Identity and organization access
Clerk authentication with application-enforced organization membership and owner, editor, reviewer and viewer roles.
- 05 / STACK
AI integration
A provider-agnostic adapter uses the OpenAI SDK with OpenAI-compatible endpoints, including configured cloud APIs and compatible local services such as Ollama or LM Studio.
- 06 / STACK
Document generation and export
The docx library creates Word files; Puppeteer with headless Chromium renders PDFs using Chrome’s HarfBuzz text shaping. ExcelJS generates XLSX files and archiver packages ZIP deliveries.
- 07 / STACK
Optional shared services
An Upstash-compatible Redis HTTP endpoint stores shared rate-limit and delivery state. Resend sends email; Sentry error reporting is enabled only when configured.
- 08 / STACK
Logging and operations
Pino provides structured application logs. The self-hosting organization manages the Node.js process, PostgreSQL, Chromium, persistent files and backups.
A starting footprint for your organization
These are conservative planning estimates, not benchmarked minimums, user-capacity guarantees or purchased support tiers. The figures below describe the application host; database and backup storage are additional.
Growing organization
Regular system-development projects across several teams or departments.
- CPU
- 8vCPU
- Memory
- 16GB
- Working storage
- 100GB SSD
Separate application and database resources; add staging, monitoring and an agreed upgrade schedule.
Large or multi-site organization
A planning baseline for larger catalogues, more complex systems and heavier export activity.
- CPU
- 16vCPU
- Memory
- 32GB
- Working storage
- 200GB SSD
Validate representative workloads before procurement. Agree data access, recovery and maintenance responsibilities with your IT team.
The current generation queue runs in one API process; larger hardware does not automatically increase generation concurrency. Multi-instance high availability needs a separate engineering review. External AI does not require a server GPU; local models need separate compatibility and hardware assessment. Allow more disk space for retained exports, uploads and local models.
Practical support, agreed before deployment
These are engagement options to discuss—not automatically included entitlements. Scope, fees, support hours, response targets and access permissions must be agreed in writing. No default 24/7 coverage or guaranteed resolution time is claimed.
Deployment assistance
Environment review, configuration guidance, installation checks and a documented acceptance checklist with your technical administrator.
Maintenance assistance
Planned upgrade guidance, compatibility checks and backup/restore review within the agreed scope.
Issue investigation
Report reproducible symptoms, the installed version and sanitized logs through the agreed contact channel. Investigation and any remote session require your approval.
Who looks after what?
Your organization
- Hosting, network security, access to the server and third-party accounts.
- Backups, monitoring, patching and an authorized IT contact.
- Approval of external AI data flows and any support access.
GlobalStandards, by agreement
- Product setup guidance and diagnosis of application issues.
- Release guidance and agreed product updates or fixes.
- Handover notes and administrator orientation within the contracted scope.
Agree these items before go-live
- 01Licence and deployment scope; ownership and intellectual property remain with Global Standards Conformity Co.
- 02Identity/domain separation, provider credentials and permitted external data flows.
- 03Representative generation and export checks, including Arabic fonts and RTL.
- 04Backup restoration, recovery responsibilities, support channels and maintenance windows.
Start with your actual environment.
Share your hosting platform, expected generation/export workload, IT contact and data restrictions. We can agree a realistic deployment and support scope.