MamboMeme

MamboMeme is a local ranked-search application for meme images and quotes. Rust will handle ingestion and the terminal interface; Python will build the search index, run lexical and semantic retrieval, rank results, and evaluate quality.

Project boundary

A query such as john cena returns ranked stored assets for the user to inspect and choose.
BM25 handles exact names, templates, quotes, and tags; dense text covers descriptions and concepts.
A Rust TUI talks to one long-lived Python retrieval worker through a versioned local protocol.
Selection feedback is optional offline evidence, never live self-training.
Context-aware suggestions reuse retrieval only after the prompt-search product works.
MMTS-Search-v1, component metrics, and contract tests justify each implemented stage.

Documentation

Architecture

Components, Rust and Python ownership, process boundaries, data flow, contracts, and planned repository shape.

Data pipeline

Source policy, parsing, canonical meme records, enrichment, deduplication, storage, publication, and pipeline health.

Retrieval

Query semantics, lexical and dense retrieval, rank fusion, filtering, result contracts, and selection feedback.

Interface

Rust terminal workflow, screen states, result previews, worker lifecycle, selection behavior, and accessibility boundary.

Testing

Contract-based unit, integration, end-to-end, security, TUI, evaluator, and performance test matrix.

Evaluation

Search benchmark, relevance and performance metrics, selection diagnostics, Technical Score, gates, and comparison protocol.

Roadmap

Evaluation-first phases for corpus processing, ranked retrieval, the Rust TUI, feedback, and later context search.

Project status

The project repository is currently a documentation-only snapshot. The corpus, implementation, model, public remote, licence, and deployment surface have not been created.

← Back