Tutorials

Build a Bento memory store in 15 minutes

Stand up Postgres + pgvector, seed three documents, and ask /v1/ask its first question.

DB2 → PostgreSQL: A Worked Example

A worked DB2 to PostgreSQL example: FETCH FIRST n ROWS ONLY, SYSIBM.SYSDUMMY1, VALUES quirks, identity columns, and SQL/PL to PL/pgSQL patterns.

Moving and assessing Oracle application code to Postgres using Swordfish

Drive the Swordfish web UI through a full Oracle to Postgres assessment of the HR demo: read the grade, find the critical PL/SQL findings, turn 336 findings into a plan.

Moving and assessing MySQL application code to Postgres using Swordfish

Drive the Swordfish terminal UI through a full MySQL → Postgres assessment of a real PHP app — find the embedded SQL, read what breaks, export the plan.

SQL Server → PostgreSQL: A Worked Example

A worked SQL Server to PostgreSQL example: TOP, IDENTITY, GETDATE/ISNULL/LEN, bracket identifiers, MERGE, and READ UNCOMMITTED habits.

MySQL → PostgreSQL: A Worked Example

A worked MySQL to PostgreSQL example: AUTO_INCREMENT, backtick identifiers, ENUM, ON DUPLICATE KEY, GROUP BY ordering, and the dump-to-PG rewrite path.

Oracle → PostgreSQL: A Worked Example

A worked Oracle to PostgreSQL example: the constructs Swordfish flags most (CONNECT BY, ROWNUM, NVL, packages), the recommended PG/EPAS change, and the human-review step.

Your first pgvector query

Install pgvector, embed three documents, and run a similarity search — in about 10 minutes.

Tutorial 0 — We Vibe-Coded a Database Doctor. Does It Actually Work?

We vibe-coded a 100-point Postgres health check. Does it actually work? The quickstart, the two engines, and an honest gut-check of where the naive version fails.

Tutorial 2 — Build an LLM-as-Judge: Catching Hallucinations Before They Ship

Build the LLM-as-judge verification layer: a cheap deterministic grounding check plus a multi-model consensus vote — including the substring-grounding bug in its natural habitat.

Tutorial 1 — Foundations: Structured Output + a Multi-Provider LLM Harness

Replace free-form LLM prose with structured Finding objects over a multi-provider harness — the checkable spine the rest of the build depends on.

Tutorial 3 — Memory & Cost: Stop Re-Diagnosing (and Re-Paying for) the Same Problems

Add cross-run memory and incremental skipping so the tool stops re-diagnosing the same issues and an unchanged re-run costs a fraction of the first.