Retrieval you can inspect—not a magic answer box.
A deterministic, browser-side lab for exploring query processing, TF-IDF ranking, thresholds, evidence selection, citations, and simple evaluation over sanitized portfolio data.
This lab does not call an LLM or send data to a server. It demonstrates the retrieval and evidence layer using a small public dataset, so every score and citation remains visible.
Run the query to synthesize an answer from retrieved evidence.
Choose a sample question to compare retrieval against its expected evidence.
Why these chunks?
Eight chunks. Clear provenance.
The corpus is intentionally small enough to inspect. Production systems need the same traceability at a much larger scale.
What changes beyond the lab.
The transparent core remains, while enterprise requirements add identity, hybrid retrieval, reranking, evaluations, observability, and controlled generation.
Ready for the architecture behind the lab?
See the public-safe case study, system map, evaluation contract, and downloadable engineering artifacts.