Open-Labor-PH
About the Project
Open-Labor-PH is an open-source initiative to structure, validate, and democratize Philippine labor law for modern AI and local language models. Decades of Department of Labor and Employment (DOLE) Orders and statutory regulations have been trapped in unstructured PDF archives, making reliable semantic querying and hallucination-free retrieval nearly impossible. Open-Labor-PH solves this by building high-throughput extraction pipelines in Rust that parse, normalize, and validate legal texts against strict JSON Schema (Draft-07) definitions, backed by cryptographic SHA-256 source manifests. An accompanying RAG evaluation benchmark suite provides checkable retrieval-augmented generation metrics across both commercial and local open-weight LLMs.
Key Features
- Decades of DOLE Orders converted from PDF into structured, schema-validated JSON
- JSON Schema (Draft-07) and cryptographic SHA-256 source verification
- RAG evaluation suite for benchmarking retrieval accuracy over legal statutes
- High-performance Rust multi-crate ingestion workspace (serde, tokio)
- Python interop via PyO3 for ML and language model evaluation workflows
- Shared open data foundation actively adopted by regional legal-tech builders
Challenges & Solutions
Legal texts are fraught with historical amendments, inconsistent layout formatting, and edge cases across decades of administrative issuances. Architecting a deterministic ingestion engine in Rust that handles PDF extraction, normalizes complex cross-references, and validates outputs against strict JSON Schemas without data loss required deep systems design. Furthermore, establishing quantitative evaluation criteria for legal question retrieval required designing ground-truth test vectors and precision metrics.
Outcomes & Impact
Open-Labor-PH is actively gaining community attention, with regional developers, legal-tech builders, and researchers reaching out to build upon the datasets. The project serves as the foundational open-research dataset for Philippine compliance, demonstrating high-performance systems engineering and applied AI data infrastructure in Rust.
Interested in Open-Labor-PH?
Let's collaborate or discuss how I can help bring your projects to life.
Get in touch