6 micro-topics in Data & Storage, each with the evidence that proves you have it and the written reason behind every link. Before you start, it rests on 5 other domains. Downstream, it holds up 2 domains.
6 topics · depth 2–12 of 16 · 5 internal links · 6 in · 3 out
Load-bearing links first. Each one says what in this domain rests on what outside it, and why.
1 link
Chunk documents for retrieval rests on Clean and normalize records
You chunk cleaned documents.
1 link
Explain embeddings rests on Explain tokenization
Embeddings operate on tokenized text.
1 link
Store and query vectors in Postgres rests on Model one-to-many and many-to-many
Vectors live in a schema alongside their metadata.
2 links
Filter retrieval by metadata rests on Query rows with SELECT
The filter half is an ordinary predicate.
Choose a vector index rests on Reason about indexes
Index tradeoffs are the same shape you met in SQL.
1 link
Generate embeddings at scale rests on Run async calls concurrently
Embedding a corpus is a batch of concurrent calls.
In prerequisite order. Each idea names the artifact that closes it, the market demand that put it on the map, and what it stands on.
Depth 2 · conceptual · unassisted · M14 §14.4
Knows what a dense vector represents
Explain tokenization · Model Foundations
Embeddings operate on tokenized text.
Depth 9 · conceptual · unassisted · M19 §19.1
Splits text so chunks are self-contained
Chunk size is chosen against what gets embedded.
Clean and normalize records · Data Ingestion
You chunk cleaned documents.
Depth 10 · procedural · scaffolded · M19 §19.1
Batches, rate-limits and stores vectors
You embed chunks.
Run async calls concurrently · Production Python
Embedding a corpus is a batch of concurrent calls.
Depth 11 · procedural · unassisted · M19 §19.2
Uses pgvector alongside relational data
Vectors must exist before they are stored.
Model one-to-many and many-to-many · Data Modeling
Vectors live in a schema alongside their metadata.
Depth 12 · conceptual · scaffolded · M19 §19.2
Trades recall against latency with IVF or HNSW
Store and query vectors in Postgres
An index is built over stored vectors.
Reason about indexes · SQL
Index tradeoffs are the same shape you met in SQL.
Depth 12 · procedural · scaffolded · M19 §19.2
Combines vector similarity with structured predicates
Store and query vectors in Postgres
Metadata filtering combines vector and relational predicates.
Query rows with SELECT · SQL
The filter half is an ordinary predicate.
Everything downstream that names an idea here as a prerequisite, grouped by where it lives.
2 links
Design a retriever-generator system rests on Store and query vectors in Postgres
Retrieval needs stored vectors.
Combine keyword and semantic search rests on Filter retrieval by metadata
Fusing results requires filtered vector search.
1 link
Persist long-term memory rests on Store and query vectors in Postgres
Semantic recall is usually vector retrieval.
Retool. Retrain. Relaunch.
375 ideas. 17 weeks. No tuition, ever.
Vets Who Code is a veteran-run 501(c)(3). The accelerator is free, remote, and we don’t take a share of your first paycheck.
Free · Remote · 17 weeks · EIN 86-2122804