Reviewer Matcher

Matches papers or proposals to suitable reviewers.

Industry	Level	Difficulty	Recommended mode
Education & research	3 of 5	Advanced	flat

Coming soon

A complete, runnable Reviewer Matcher reference implementation is in progress. The architecture and core code below already work with SrvDB 0.3.

The problem

• Finding reviewers is slow.
• Conflicts and mismatches happen.
• Expertise is described in prose.

How SrvDB powers Reviewer Matcher

1. Reviewers’ publications are embedded.
2. Submissions query the closest experts.
3. Scores rank candidates.
4. Pools updated with upsert.

What changes

Today	With Reviewer Matcher on SrvDB
Manual reviewer search	Ranked candidates
Mismatched reviews	Better matches
Static pools	Updated pools

How it works

1. Embed reviewer profiles.
2. Embed submissions.
3. Rank matches.
4. Check conflicts.
5. Assign.

SrvDB features it uses

• Exact search (flat): every query compares against every stored vector, so results are exact and explainable.
• True cosine scores: scores in [-1, 1] mean the same in every mode, so thresholds are stable.
• search_batch: runs many queries at once, in parallel inside Rust.
• Crash-safe persist(): after it returns, data survives a crash or power loss and reopens exactly as committed.

Use cases

• Journal reviews
• Conference committees
• Grant panels
• Thesis examiners
• Internal reviews
• Mentor matching

Core code

embed() is your embedding model (for example sentence-transformers, or a function that turns your records into feature vectors).

import json
import srvdb

db = srvdb.Database("./reviewer-matcher-db", dimension=768)

def ingest(items):
    """items: list of (id, text, metadata dict)"""
    db.upsert(
        ids=[i for i, _, _ in items],
        embeddings=[embed(t) for _, t, _ in items],
        metadatas=[json.dumps(m) for _, _, m in items],
    )
    db.persist()                      # crash-safe checkpoint

def find_reviewers(submission_abstract, k=5):
    hits = db.search(embed(submission_abstract), k)
    return [(json.loads(db.get(i)), round(score, 3)) for i, score in hits]

Limits and honest notes

• Conflict-of-interest checks remain manual.

Build roadmap

1. Prototype: load a sample of real data with flat mode and check that searches return what an expert expects.
2. Measure: if you move to an approximate mode, compare its results with flat on the same queries (recall).
3. Pipeline: add continuous upsert and decide when to call persist().
4. Harden: test restarts and crashes, add logging, review every output path with domain experts.
5. Deploy: ship the database directory with the application or service.

All Education & research projects

Level	Project	What it does
1 · Beginner	Paper Search	Search a personal or lab library of papers by meaning.
2 · Intermediate	Study Assistant	Answers questions only from the student’s own course material.
3 · Advanced	Reviewer Matcher (this page)	Matches papers or proposals to suitable reviewers.
4 · Expert	Research Group Brain	A shared, searchable memory for a lab: notes, data descriptions, figures.
5 · Frontier	Scientific Similarity Lab	Similarity search over spectra, molecules and other scientific data.

See also: Use cases · Live examples (coming soon)