Self-Hosted Odoo Private AI Layer Blueprint Generator
This free generator drafts a private AI architecture plan for self-hosted Odoo. It builds on the external API and record rules, adds attachment retrieval and a local model, and exports a shareable PDF or Word document.
Fill in
Self-Hosted Odoo Private AI Layer Blueprint Generator
Private AI Layer Architecture Plan: Self-Hosted Odoo
Company: Northfield Machining Ltd System and version: Self-Hosted Odoo, Odoo 17 Database: PostgreSQL 15 Hosting: On-prem server Named users: 80 Sites or plants: 2 GPU: 1-2 x 48 GB class GPUs Target go-live: 2027-04-30
Context and objectives
Northfield Machining Ltd wants an AI assistant over its Self-Hosted Odoo data and documents that runs inside its own environment (On-prem server). The assistant must answer questions in natural language, show the data behind each answer, respect existing user permissions and remain read-only until approvals are in place.
Main use cases: Manufacturing order status and component shortages. Sales order and delivery questions. Search of vendor documents attached to purchase orders.
Data boundaries
The following data must never leave the site and must stay out of any external service: Customer pricelists. Employee and payroll data. Controlled drawings.
All processing, indexing and model inference run on infrastructure controlled by Northfield Machining Ltd.
Components
| Layer | Design | Notes for Northfield Machining Ltd |
|---|---|---|
| Data access path | External API (XML-RPC, JSON-RPC or the newer JSON API in recent versions) called as the signed-in user so access rights and record rules apply; optional PostgreSQL replica for aggregates only | Check which endpoints Odoo 17 supports |
| Retrieval index | Vector index of ir.attachment files and knowledge articles tagged with res_model and res_id; read access to the parent record checked before use | Filestore location needed from IT |
| Model serving | Open-weight model on vLLM or Ollama sized to 1-2 x 48 GB class GPUs | Isolated host when hosting is On-prem server |
| Tool server | Typed read tools wrapping search_read calls with fixed fields and limits | Tools respect pagination on large tables |
| Identity and permissions | Per-user API keys; multi-company context passed on every call; no shared admin key | Direct SQL never used for per-user answers |
| Audit | Append-only log of question, user, model, domain filter and result count | Reviewed in the pilot |
Security controls
| Control | Implementation | Driven by |
|---|---|---|
| Network isolation | Model, index and tool server in a dedicated zone with no outbound internet route | Hosting: On-prem server |
| Data minimisation | Only approved views and document sets are exposed; excluded data is never indexed | Data that must never leave the site |
| Least privilege | Read-only service identity; per-user filtering of tool calls | 80 named users |
| Model governance | Open-weight model pinned by version, weights checksummed, updates via change control | GPU: 1-2 x 48 GB class GPUs |
| Audit and retention | Append-only log of questions, tools, sources and users with an agreed retention period | Internal audit and compliance |
| Write-back control | No writes in phases 1 and 2; later writes only through supported APIs with named approvers | Risk appetite of Northfield Machining Ltd |
Rollout plan
| Phase | Scope | Exit criteria |
|---|---|---|
| 1. Read-only questions | Approved views and documents for a pilot group; answers show their sources | Accuracy sampled by owners; no permission leaks in review; log reviewed weekly |
| 2. Drafts | Assistant drafts emails, summaries and suggested changes for a person to copy | Users edit less over time; no unreviewed output leaves the system |
| 3. Approved actions | Selected actions executed through supported APIs after human approval | Approver named per action; rollback tested; go-live target 2027-04-30 |
Open questions for IT
1. How fresh must answers be: real time, minutes or daily? 2. Which companies, databases or libraries are in scope across 2 site(s)? 3. Who owns document permissions and how are they mapped to Self-Hosted Odoo roles? 4. Which customisations or custom tables must be exposed? 5. What is the backup, monitoring and patching plan for the GPU host? 6. Who approves any write-back and through which supported interface? 7. Is a pilot group of 10 to 20 users available for phase 1?
Approvals
Starting-point document, not a final design. Confirm table names, interfaces and security behaviour for your exact release with your vendor or implementation partner. Nothing you enter leaves your browser.
Optional Odoo architecture review
We can review the plan against your Odoo version, modules and security rules.
Optional. The tool and every export work without this. No spam.
How to use the blueprint
Choose your Odoo version, PostgreSQL version and hosting, then list the questions people ask about manufacturing, sales, stock and purchasing. Note any custom modules and the data that must not leave your servers. The plan describes how an assistant authenticates as each user, reads through the ORM, retrieves attachments and calls a local model, with a three-phase rollout.
What Odoo specifics it assumes
The plan assumes a self-hosted Odoo, because Odoo Online and Odoo.sh are cloud hosted. It relies on the ORM to enforce ir.model.access groups and ir.rule record rules, and on per-user API keys. A PostgreSQL replica is optional and limited to aggregates, because direct SQL bypasses record rules. API endpoints differ by version, so the plan asks IT to confirm which ones apply.
Reading the output and common mistakes
Use the open questions to confirm multi-company behaviour, custom models and filestore access. Typical mistakes are sharing one administrator key, using SQL for per-user answers, unbounded search_read calls and indexing attachments without checking parent access. Tailor the tool list to the apps you actually run.
Frequently Asked Questions
Does this work with Odoo Community?
Yes. The external API exists in Community and Enterprise when self-hosted. The plan lists tools by app, so include only the modules you run.
Why per-user API keys?
Calls made with a person's key run through Odoo's ORM as that person, so access rights and record rules apply automatically. A shared admin key would let the assistant show data users cannot normally see.
Does it work on Odoo.sh?
The plan targets self-hosted installs. Odoo.sh and Odoo Online are hosted by the vendor or a cloud provider, so a fully on-prem model and index are not the same pattern.
Is Odoo's own AI enough?
It might be, depending on version and data policy. This plan is for teams that need data to stay on their servers or want one assistant across Odoo and other sources.
If you want help adapting this plan to your Odoo modules and customisations, Netray can review it with you. The tool works without us.
Related Tools
Self-Hosted Odoo AI Assistant Sizing and Value Calculator
Estimate GPU count, token volume, index size, hours saved and payback for a private on-prem AI assistant on Self-Hosted Odoo.
Developer ToolsOdoo XML-RPC BOM Export Python Script Generator
Fill in your Odoo URL, database and filters to get a ready Python script that exports BOMs and component lines to a CSV file.
Other ERPsOdoo Manufacturing Technical Model Names Reference
Searchable lookup of the Odoo models behind manufacturing, inventory, purchasing and sales, with the fields you query most.
On-Prem AIOn-Prem AI Security Hardening Checklist
A practical control checklist for securing self-hosted language models, covering model provenance, network isolation, data governance, host hardening, and audit readiness.
More Developer Tools tools
See all- SAP Sales Order ABAP SELECT Generator (VBAK, VBAP, VBFA)
- SAP Sales Order Create BAPI ABAP Generator (BAPI_SALESORDER_CREATEFROMDAT2)
- SAP ABAP Open Items SELECT Generator
- SAP ACDOCA SQL Query Generator
- SAP QM Inspection Lot ABAP SELECT Generator
- SAP Quality Notification Pareto SQL Query Generator
- SAP QM Usage Decision BAPI Skeleton
- ABAP Custom Exception Class Generator
- ABAP Unit Test Class Generator
Keep exploring
Free tools
- Self-Hosted Odoo AI Assistant Sizing and Value CalculatorEstimate GPU count, token volume, index size, hours saved and payback for a private on-prem AI assistant on Self-Hosted Odoo.
- Macola Private AI Layer Blueprint GeneratorGenerate a shareable architecture plan for adding a private, on-prem AI assistant to Macola, with components, security controls and a three-phase rollout.
- Odoo Manufacturing Setup ChecklistA practical checklist for configuring Odoo Manufacturing from inventory foundations through BOMs, work centers, planning and go-live.
ERP fixes and how-tos
- Self-Hosted Odoo AI Assistant: Private Architecture With a Local ModelAdd a private AI assistant to self-hosted Odoo: external API with user API keys, record rules respected, PostgreSQL replica, local model and audit trail.
- Odoo Manufacturing Reports: Production Analysis, Cost and ForecastBuilt-in Odoo manufacturing and production reports, how to forecast demand and component needs, and how to build custom MRP reports with SQL or Studio.
- Odoo Manufacturing Tutorial: From BOM to Finished ProductOdoo manufacturing tutorial for small manufacturers: set up products, BOMs, work centers and operations, run a manufacturing order, and what to check in Odoo 19.
AI on your ERP
- AI for Odoo Manufacturing That Reads Your Own Database, Not a Vendor's CloudAdd grounded, natural-language AI to Odoo Manufacturing without sending MRP, BoM, or work order data to a public model. On-prem or private cloud, ORM-native.
- AI for Odoo Manufacturing (MRP): BOMs, Work Orders and ShortagesAI for Odoo Manufacturing: BOM checks, work order delays, reordering rules and plain-language questions over mrp.production data, on-prem or private cloud.
Integrations
- How to add AI agents to OdooHow to add AI agents to Odoo Manufacturing: a tool layer over the external API, approval before writes, and agents for MRP exceptions, POs and quality.
- How to connect ChatGPT to OdooHow to connect ChatGPT to Odoo: custom GPT actions, exports versus live access, enterprise data controls, and when a private model is the safer choice.
Software
- OdooWhat Odoo is, how its open-source Community and Enterprise editions differ, how the manufacturing apps and XML-RPC API work, and which alternatives to compare.
- SAP Business OneSAP Business One explained for small manufacturers: modules, Service Layer and DI API, deployment, limits, alternatives and where AI helps on top of it.