Delhi Metro Text Map, Context-Engineered for LLMs (github.com)

The “Delhi Metro Text Map — Context‑Engineered for LLMs” is a plain‑text, LLM‑friendly transit dataset and README you can paste into ChatGPT (or prompt by searching its name) to get grounded answers about Delhi Metro routes. It enumerates every line with station IDs, ordered stops, latitude/longitude coordinates, river crossings, branch vs main distinctions, and explicit interchange mappings including platform numbers and which platform connects to which line. The file even gives usage tips (paste the README into your prompt) so an LLM has the full route graph as context rather than guessing. For AI/ML practitioners this is a compact example of schema‑aware context engineering and a lightweight knowledge base for retrieval‑augmented generation: it supports precise route planning, nearest‑station queries, transfer counting, and geospatial reasoning while reducing hallucinations. Technically notable details include unique station tokens (e.g., [R16]=Kashmere Gate), geo coords for spatial distance computations, explicit platform links for multi‑leg routing, and branch/main line semantics (Blue main vs branch). Practical next steps include converting it to a graph/JSON, embedding it for vector search, or integrating with a simple planner for shortest‑transfer or distance‑based routes. Caveats: it’s a static snapshot (no live schedules or disruptions) and benefits from light preprocessing to validate coordinates and normalize naming.