PbGraphRAG Developer Challenge – Legal Document Processing (Prototype) /b /pulliRole: Senior RAG Systems Developer (Contract/Freelance) /liliCompensation: $600 — paid only if you pass (95% benchmark) /liliTimeline: 3–5 days from materials receipt to live demo /liliPurpose: Technical evaluation for potential long-term hire /liliFrontend/UI: None (backend prototype only) /liliContact: /li /ulpbObjective /b /ppWe’re seeking an expert in bgraph-based retrieval (GraphRAG) /b to build a high-accuracy prototype for legal document reasoning. This is a paid technical test that may lead to a long-term position. The goal is a true GraphRAG system featuring explicit knowledge-graph construction and traversal, multi-hop reasoning, agentic orchestration, and strong focus on retrieval accuracy and explainability. /ppbMaterials Provided /b /pulli/docs/ → Pre-processed Markdown legal documents with metadata /lili/sample_questions.json → Example question format /lili/sample_answers_rag.json → Example answer format /li /ulpDownload materials: /pp (Benchmark uses unseen questions.) /ppbDeliverable /b /ppImplement two functions in bPython 3.12 /b (Poetry project): /ppdef ingest(document_paths: List(str)) - None: /pp """Ingest Markdown docs and build the knowledge graph.""" /ppdef query(questions: List(str)) - List(str): /pp """Return answers with Vancouver-style citations grounded in retrieved sources.""" /ppbr/ppbRequirements: /bNo UI, no API keys provided. Any stack may be used. query(...) must support parallel execution (~400 questions in ≤60 min) and show a progress indicator. Test thoroughly for correctness and performance before the demo. /ppbLive Demo /b /ppIn a 60-minute live session you will: /pulliReceive ~400 unseen questions. /liliRun query(...) to produce /answers.json. /liliExplain your architecture: how the graph is built, traversed, and used to generate grounded answers. /li /ulpOnly the bdeveloper(s) /b who wrote the code may present. /ppbEvaluation /b /ppPassing requires an overall score above 95%, measured by (LLM as a judge): Faithfulness (grounded, no hallucinations), Relevance (retrieval matches intent), Completeness (covers key legal points), and Clarity (structured, legally coherent writing). /ppbPayment Next Steps /b /ppIf you pass, you’ll receive $600 USD after verification of reproducibility and hand-over of the repo (codebase, Poetry lock, run instructions, brief tech note). Top performers may be invited to interview for a long-term paid role. Failure to pass or complete within 60 minutes = no payment (you keep your code). /ppbKey Priorities /b /ppParallelization, graph-based reasoning, correctness, and explainability. /ppbNote for Agencies /b /ppWe will not engage in pre-contract discussions with agencies. If an agency wishes to propose a developer, communication will proceed only after that developer passes the benchmark test. This ensures time efficiency and direct technical validation. /p