Tips and Tricks #221: Use Embeddings for Semantic Search
Implement semantic search using text embeddings for more relevant results than keyword matching.
Designing intelligent systems, one layer at a time.
Implement semantic search using text embeddings for more relevant results than keyword matching.
Build modular, tested, documented data transformations with dbt.
Use table partitioning to dramatically speed up queries on large datasets.
Use MERGE (upsert) for safe, rerunnable data pipelines that handle duplicates gracefully.
Calculate running totals, rankings, and moving averages efficiently with SQL window functions.
Eliminate heap allocations when parsing strings by using Span for memory-efficient operations.