Skip to content

Latest commit

 

History

History
33 lines (24 loc) · 1.36 KB

File metadata and controls

33 lines (24 loc) · 1.36 KB

Project Instructions for Codex Agents

This repository is a research benchmark for Romanized Nepali customer-support retrieval-augmented generation. It is not a production chatbot.

Scope Control

  • Keep baseline systems simple and controlled.
  • Phase 1 is only Traditional Semantic RAG.
  • Do not mix Phase 2 Agentic RAG behavior into Phase 1.
  • Do not mix Phase 3 Agentic GraphRAG behavior into Phase 1.
  • Do not add query rewriting, intent routing, verifier agents, graph extraction, or graph lookup to Phase 1 code.

Reproducibility

  • Preserve deterministic data preparation and chunking where possible.
  • Always save experiment outputs under results/.
  • Use explicit CLI arguments for experiment variation.
  • Do not silently ignore missing input files.
  • Keep output schemas stable so systems can be compared fairly.

Secrets and Configuration

  • Do not hardcode API keys or provider secrets.
  • Read LLM provider keys and model names from the environment.
  • Supported Phase 1 LLM providers are Google Gemini and NVIDIA NIM only.
  • The benchmark must support retrieval-only runs when no LLM API key is available.

Engineering Style

  • Prefer small, testable modules.
  • Prefer readable research code over framework-heavy production abstractions.
  • Keep dependencies focused on the current phase.
  • Add tests for loaders, schemas, retrieval structures, and output writing.