Developer Guides¶

Practical guides for building applications with Uni.

### [Cypher Querying](cypher-querying.md) Complete OpenCypher reference with pattern matching, filtering, and aggregations.

### [Temporal Intervals](temporal-intervals.md) BTIC temporal intervals with granularity, certainty, and Allen's interval algebra.

### [Vector Search](vector-search.md) Semantic similarity search with HNSW and IVF_PQ indexes.

### [Data Ingestion](data-ingestion.md) Bulk import, streaming writes, and ETL patterns.

### [Schema Design](schema-design.md) Best practices for modeling your data as a property graph.

### [Pydantic OGM](pydantic-ogm.md) Type-safe Python models with automatic schema generation.

### [Performance Tuning](performance-tuning.md) Optimization strategies for queries, indexes, and storage.

### [Sparse Vectors](sparse-vectors.md) Learned-sparse (SPLADE) retrieval: schema, indexing, and `uni.sparse.query`.

### [BGE-M3 Hybrid Retrieval](bge-m3-hybrid-retrieval.md) Dense + lexical + learned-sparse fusion end to end with BGE-M3.

### [AI Skill](ai-skill.md) Using the packaged `uni-db` skill to drive coding agents against this API.

Common Tasks¶

MATCH (p:Paper)-[:CITES]->(cited:Paper)
WHERE p.year > 2020
RETURN cited.title, COUNT(*) as citations
ORDER BY citations DESC
LIMIT 10

CALL uni.vector.query('Paper', 'embedding', $query_vector, 10)
YIELD node, score
RETURN node.title, score

uni import semantic-scholar \
  --papers papers.jsonl \
  --citations edges.jsonl \
  --output ./storage