Odoo Vector Search
Find exactly what you need in seconds. Our vector search technology creates semantic embeddings of Odoo source code, documentation, and community knowledge for instant, context-aware retrieval.
Stop scrolling through Stack Overflow. Get precise answers to your Odoo development questions with semantic understanding of intent, not just keyword matching.
Try Vector SearchWhat Is Vector Search for Odoo?
Traditional keyword search fails when you do not know the exact terms. Vector search understands the meaning behind your query.
Traditional Keyword Search
- ✕Search “how to add discount to sales” — returns pages about the word “discount”
- ✕Cannot find code patterns unless you know the exact method name
- ✕Misses synonyms: “rebate” and “discount” are treated as completely different
- ✕No understanding of Odoo-specific context or framework semantics
Letzdoo Vector Search
- Search “how to add discount to sales” — finds sale.order price computation code
- Finds semantically similar code patterns across entire Odoo source
- Understands intent: finds discount, rebate, price reduction, and promo implementations
- Odoo-specific embeddings understand ORM, views, and framework conventions
How Vector Search Works
A three-stage process that transforms code and documentation into searchable semantic embeddings.
Code Chunking & Analysis
Your Odoo codebase is intelligently chunked into semantic units — not arbitrary line counts. Models, methods, views, and configuration files are parsed with Odoo-aware AST analysis to preserve meaningful boundaries. Each chunk retains metadata about its module, model, and purpose.
Embedding Generation
Each code chunk is transformed into a high-dimensional vector embedding that captures its semantic meaning. We use specialized models trained to understand code semantics, so a Python method that computes discounts is close in vector space to XML views that display discount fields — even though they share no keywords.
Semantic Retrieval
When you search, your query is embedded into the same vector space and matched against your codebase using cosine similarity. The most semantically relevant code, documentation, and patterns are returned instantly — providing Claude AI with exactly the context it needs to generate accurate, codebase-aligned solutions.
What Gets Indexed
Your Custom Modules
Every model, view, controller, wizard, and configuration file in your Odoo addons path.
Odoo Core Source
The complete Odoo source code for your target version, including base models, core views, and framework internals.
Official Documentation
Odoo developer documentation, API references, and migration guides for all supported versions.
Community Patterns
50,000+ validated code patterns from the OCA (Odoo Community Association) and top-rated community modules.
Best Practices
Curated implementation patterns, security guidelines, and performance optimization strategies.
Industry Solutions
Vertical-specific implementation patterns for manufacturing, retail, services, and more.
Vector Search Performance
Frequently Asked Questions
What is vector search and how does it apply to Odoo?
Vector search converts code and documentation into mathematical embeddings that capture semantic meaning. Unlike keyword search which matches exact terms, vector search understands concepts — searching 'customer invoice workflow' finds relevant code even if those exact words don't appear. For Odoo, this means instantly finding relevant modules, methods, and patterns across massive codebases.
How fast are vector search queries on Odoo codebases?
Our vector search returns results in under 100ms, even across codebases with 50,000+ files. The embedding index is pre-built during analysis, so queries are nearly instant. This compares to minutes or hours of manual searching through Odoo module directories and XML views.
What types of Odoo content can be searched with vectors?
Everything: Python model definitions, XML views, QWeb templates, JavaScript/OWL components, CSV data files, manifest configurations, and documentation. The system also indexes relationships between modules, inheritance chains, and field dependencies to provide contextual results.
How does vector search differ from grep or IDE search?
Traditional search tools match exact text patterns. Vector search understands intent — you can search for 'how to add a custom field to sale orders' and find the exact model extension pattern, even if no file contains that exact phrase. It also ranks results by relevance and can find similar patterns across different Odoo modules.
Can I use vector search with my private Odoo repositories?
Yes. Your code is embedded in an isolated vector space accessible only to your team. Embeddings are generated using secure, privacy-preserving models and stored with Supabase RLS policies. We never mix your code embeddings with other customers' data, and source code is never stored — only the mathematical representations.
Search Your Odoo Codebase Semantically
Index your Odoo modules and start finding answers in milliseconds instead of hours.
Start Searching