Our proposed database review workflow: static SQL rules enforce policy, Jev assesses contextual risk, an LLM explains the findings, and a human approves.

Testing Jev on Database Change and Access Workflows

Oracle SQL review tools compared: PLSQL_WARNINGS, SQLcl CODESCAN, dbLinter, SonarQube, SQLFluff, Toad, Flyway, utPLSQL, and where Bytebase's approval gate fits.

Top Oracle SQL Review Tools in 2026

Map key SOC 2 criteria in CC6, CC7, and CC8 to example database access controls and audit evidence, including who can read production data.

Database Access Control for SOC 2: CC6, CC7, and CC8 at the Database

Compare zero-downtime schema migration strategies: native online DDL, expand-and-contract, online schema change tools, and blue-green deployments.

Zero-Downtime Database Schema Migration Strategies

A SQL Server schema migration tool moves a database from one schema version to the next. This guide reviews 8 tools, migration-based and state-based.

Top 8 SQL Server Schema Migration Tools in 2026

Replit's agent wiped a production database during a code freeze, then said rollback was impossible. It wasn't. Every step of that failure maps to a control you can put in place today.

How to Prevent an AI Agent From Dropping Your Production Database

Streaming production to a warehouse and masking in dbt masks for analysts, not operators. Where cleartext survives, and what production masking needs.

Data Masking Architecture: Why the Warehouse Pattern Leaves Production Exposed

SQL Server SQL review tools compared: the built-in SR rules in SQL projects, SqlServer.Rules' 140+ T-SQL checks, SQLFluff, tSQLt, Redgate, and where Bytebase's approval gate fits.

Top SQL Server SQL Review Tools in 2026

A SQL Server schema compare tool diffs two databases and generates the T-SQL to reconcile them. This 2026 guide reviews 8 tools and what each one costs.

Top 8 SQL Server Schema Compare Tools to Diff and Sync Databases in 2026

What the HIPAA Security Rule asks of database access: a name on every session, minimum necessary at the column, emergency access that still gets logged, and an audit trail that can say who read a patient's chart. The controls behind each, what OCR fines, and what the proposed 2027 rule tightens.

Database Access Control for HIPAA: The Security Rule at the Database

I followed Flyway's, Liquibase's, and Bytebase's own recommended quickstarts to a first managed PostgreSQL migration. Two took about half an hour. One took about an hour and a half.

From Zero to First Migration: How Much Setup Do Flyway, Liquibase, and Bytebase Require?

MySQL SQL review tools compared: SQLFluff for style, Skeema for schema-design linting, SOAR for query anti-patterns, and where Bytebase fits as the approval layer.

Top MySQL SQL Review and Lint Tools in 2026

What PCI DSS v4.0.1 asks of database access: least privilege by role, MFA on every CDE session, service accounts treated like humans, masked PANs, and a query-level trail kept for a year. Where each control lives, and where Bytebase stops.

Database Access Control for PCI DSS: Requirements 7, 8, and 10 at the Database

How PostgreSQL evolved from blocking DDL toward concurrent operations, metadata-only changes, and deferred constraint validation.

The Evolution of PostgreSQL Schema Migrations: Less Locking, Less Rewriting, Less Downtime

RDS IAM authentication replaces a stored database password with a 15-minute token. This reference implementation adds the rest of the JIT path: a scoped IAM role, verified identity, a bounded session, approval in Git, and an honest look at the four limits that remain.

Just-in-Time Database Access with AWS RDS IAM Authentication

Four missing primitives: delegation, attenuation, immutable application context, and labels on data. GRANT and RLS are a good policy engine, but nothing trustworthy feeds it when the caller is an AI agent. Oracle and SQL Server solved pieces of this; Postgres has none of it.

What Postgres is Missing for AI Agents

Bytebase and Delphix both mask sensitive data, but at different points: Delphix masks non-production copies, Bytebase masks live production queries.

Bytebase vs. Delphix: a tale of two data masking stories

Comparing HashiCorp Vault, Infisical, Doppler, and 1Password, the vendor-neutral options for database credentials that work across clouds and on-prem.

Best Secrets Manager for Database Credentials in 2026: Vault vs Infisical vs Doppler

On-behalf-of (OBO) is the pattern for AI agents to act with a caller's delegated identity instead of a shared credential. No mainstream database engine carries it the way agents need, and here is why a governance layer in front of the database is where it has to live.

On-Behalf-Of Database Access for AI Agents

Postgres SQL review tools compared: SQLFluff for style, Squawk for migration safety, plpgsql_check and pgTAP, and where Bytebase's rule engine overlaps them with an approval gate attached.

Top Dynamic Data Masking Tools in 2026

Compare direct database CLI access with a governed MCP server across six dimensions: setup, reach, identity, governance, auditability, and workflow fit.

MCP vs. CLI: How an AI Agent Should Reach Your Database

Zero trust for databases extends identity and network controls to SQL statements, query results, data changes, and schema changes.

Zero Trust for Databases: From Access Control to SQL Control

Snowflake and the data warehouses can tag any database object with structured metadata and drive masking from it. Almost no production database can. In the LLM era, that inline, self-contained context is exactly what agents need.

Object Tagging: The AI Feature Almost Every Production Database Skipped

DbUp runs your .NET migrations. Bytebase governs the database: change, access, and compliance in one platform. A practitioner comparison.

Bytebase vs. DbUp: from migration runner to unified governance

Database governance lives upstream, on the production database. Here is how to actually govern the source: reviewed changes, scoped access, codified policy, and audit where the data is born.

How to Govern Data at the Source: A Guide to Upstream Database Governance

Data governance and database governance are four letters apart and constantly conflated. They are different disciplines, owned by different teams, solved by different tools. Here is the dividing line.

Data Governance vs. Database Governance: The Difference Is the Base

PopSQL shuts down on September 1, 2026. Compare Bytebase and PopSQL across query collaboration, supported databases, access control, data masking, audit logging, and pricing before you migrate.

Bytebase vs. PopSQL: a side-by-side comparison for migrating teams

PopSQL shuts down on September 1, 2026. Compare the top PopSQL alternatives for team SQL editing, dashboards, and governed database access before you migrate.

Top PopSQL Alternatives 2026: Where to Migrate Before the Shutdown

The best multi-database MCP servers in 2026. Compare DBHub, Google's MCP Toolbox for Databases, and Bytebase on engine breadth, host neutrality, and governance for connecting AI agents to more than one database. Single-engine Postgres servers are covered in a companion post.

Top 3 Open Source Multi-Database MCP Servers in 2026

The best open source Postgres MCP servers in 2026. Compare Postgres MCP Pro, Supabase MCP, and pgconsole on host neutrality, access model, and governance for connecting AI agents to PostgreSQL. Multi-database MCP servers are covered in a companion post.

Bytebase vs. SchemaHero: a side-by-side comparison for database schema migration

Money software is judged not on whether the feature works, but on what the database can prove eighteen months later. Fintech database governance through three principles: no invented data, no lost data, no trust.

Database Governance for Fintech: Guarding the Ledger

PostgreSQL has no ADD COLUMN ... AFTER. The reason one little number in a system catalog turned a "simple" feature into a decade-long engineering saga, and what you can do about it today.

Why It's So Hard to Add a Column in the Middle of a PostgreSQL Table

What a database MCP server is, what it hands an AI agent, where the credentials live, and why a raw one is fine locally but needs access control against production.

What is a Database MCP Server?

Most database MCP servers hand an agent a connection string and call it a day. The MCP server is the one place access control, masking, approval, and audit either apply to the agent or get walked around.

Governed MCP vs. Raw MCP: How Agents Reach Your Database

AI agents are no longer just reading your database, they are authoring the migrations. The write path is a different risk class, and it needs its own gate.

When the Agent Writes the Migration, Who Approves It?

Generic PAM controls the database connection. Database PAM controls the SQL statement: per-query authorization and approval, SQL review, data masking, and statement-level audit. What the category means and why it exists.

What is Database PAM (Privileged Access Management)?

How to run SQL Server schema migrations safely in production, covering state-based vs migration-based workflows, online index operations, batched backfills, the expand-and-contract pattern, Availability Groups, and CI/CD.

SQL Server Schema Migration and Change Management

CyberArk is a broad privileged access management (PAM) platform. Bytebase is a database-specific PAM for access control, masking, and SQL review. If you already run CyberArk, here is why you still need it for the database.

Bytebase vs. CyberArk: the database vertical in your PAM stack

Database credentials management best practices: store passwords and secrets in one place, rotate on a schedule, and separate the application path from the human path.

Database Credentials Management: Best Practices

The practical ways to audit MySQL activity, from the general query log to audit plugins, binlog CDC, and Bytebase, and where each one falls short.

MySQL Audit Logging Guide

Every critical asset has its system of record, except the database. Bytebase is the one. The standard for database development.

The Standard for Database Development

Dynamic data masking is chosen by access path, not product. Five approaches (engine-native, in-database, application, BI layer, and Bytebase) and which caller each one masks.

How to Implement Dynamic Data Masking: Five Approaches by Access Path

Static vs dynamic data masking, when each applies, how they differ on backups, performance, and reversibility, and when to run both.

Static vs Dynamic Data Masking: When to Use Which

The static data masking tools that produce safe non-production data, Delphix, Tonic.ai, and greenmask compared, plus how to roll your own when no tool fits.

Static Data Masking Tools (and How to Roll Your Own)

Compare BigQuery data masking options: native column-level dynamic data masking (built-in masking rules via policy tags and data policies, Dataplex catalog) and Bytebase fleet-wide masking.

BigQuery Dynamic Data Masking

Compare SQL Server data masking options, native Dynamic Data Masking (five mask types, SQL Server 2022 granular UNMASK) and Bytebase fleet-wide masking.

SQL Server Dynamic Data Masking

Compare Oracle data masking options. Native Data Redaction (five function types via DBMS_REDACT, EXEMPT REDACTION POLICY) and Bytebase fleet-wide masking.

Oracle Dynamic Data Masking

You write a one-line UPDATE against production. Here's the lane that keeps it safe: the four phases, the controls, and what each one prevents.

How to Handle Ad-hoc Database Data Changes?

Notes on the MySQL 9.7 LTS release: Hypergraph Optimizer in Community, JSON Duality View DML, Group Replication observability, and Dynamic Data Masking for Enterprise.

Notes on the MySQL 9.7 LTS release

Most DAC implementations stop at the connection. Real DAC operates at the statement and covers three vectors: machine, human, and AI agent.

What is Database Access Control (DAC)?

How to run dynamic data masking so it survives an audit: mask by role, go column-level, log unmask events, review policy changes in code, and keep environments in sync. Plus a vendor comparison and an end-to-end walkthrough.

Dynamic Data Masking Best Practices

How Bytebase audits the workflow layer: every SQL, schema change, approval, and export tied to a real user identity, with structured exports for any SIEM.

How Bytebase Handles Audit Logging

PostgreSQL 19 features I'm excited about: planner advisor framework, DDL extraction, online REPACK, pg_stat_recovery, parallel autovacuum, sequence sync, and 64-bit MultiXact.

PostgreSQL 19 features I'm excited about

Bytebase is HIPAA-compliant — independent verification that our administrative, physical, and technical safeguards meet the HIPAA Security Rule for handling Protected Health Information.

HIPAA Compliance

PAM governs who can reach a database. Bytebase governs what SQL and schema changes happen once they get there. Where they differ, and why teams use both.

Bytebase vs. PAM: Why Database Teams Need More Than Privileged Access Management

What SOC 2 compliance audits actually want from your database audit log: not 'logging enabled' but evidence your auditor will accept. Lessons from going through our own audit.

SOC 2 Audit Log Requirements: Lessons From Our Own Audit

Bytebase vs. Atlas compared across migration approach, developer interface, supported databases, ORM integration, linting, approval, access control, audit, and pricing. Updated for Atlas v1.3 and Bytebase 3.22.

Bytebase vs. Atlas: a side-by-side comparison for database schema migration

Database access control practices that hold up in a SOC 2 audit: least privilege, separation of duties, just-in-time access, statement-level enforcement, and the common mistakes that get flagged.

Database Access Control (DAC) Best Practices

A practical look at what SOX Section 404 expects from database changes: the four ITGC domains, the six questions auditors actually ask, and how to make compliance evidence a byproduct of your workflow.

Database Change Management for SOX Compliance

TRUNCATE and DELETE both remove rows, but they differ in speed, rollback behavior, and locking. Here is how each one behaves in PostgreSQL, MySQL, and SQL Server, and when to reach for which.

TRUNCATE vs DELETE in SQL: Differences Across PostgreSQL, MySQL, and SQL Server

A practical walk through six database design patterns you will actually meet in production: normalization, star schema, EAV, polymorphic associations, multi-tenancy, and the anti-patterns worth avoiding.

Database Design Patterns Every Developer Should Know

A practical walkthrough of pgvector: installing the extension, storing embeddings, writing similarity queries, building HNSW indexes, and knowing when plain PostgreSQL is enough.

pgvector: How to Add Vector Search to PostgreSQL

How upsert works across MySQL, PostgreSQL, and SQL Server, with the syntax for each, the edge cases worth knowing before you ship, and how upserts interact with migrations.

Upsert in SQL: INSERT ON DUPLICATE KEY vs ON CONFLICT Explained

AI agents need more than DDL. The full database lifecycle codified (schema, dependencies, ownership, history) gives LLMs the context to query correctly.

From Schema as Code to Schema as Context

How to find and fix slow MySQL queries: reading the slow query log, using EXPLAIN, tuning indexes, and rewriting queries, with a worked example for each.

How to Fix Slow MySQL Queries: A Practical Guide

Convex and Supabase are both backend-in-a-box platforms, but built on different foundations. This post compares their architecture, real-time model, data model, schema management, and pricing to help you decide.

Convex vs Supabase: Which Backend Platform Should You Choose?

Blue-green deployment is easy when your servers are stateless. The trouble starts at the database. This guide walks through why database blue-green is hard and how to do it without corrupting your data.

Database Blue-Green Deployment: A Practical Guide

A look at 10 data management tools developers and data engineers actually use in 2026: GUI clients, schema migration, pipelines, and analytics, with pricing for each.

Top 10 Data Management Tools in 2026

How to give AI agents governed access to enterprise data: access control, Dynamic Data Masking, audit logging, and just-in-time grants, applied to a new principal type.

How to Govern AI Agent Database Access

The best SQL editors built exclusively for PostgreSQL: what each one is good at, who it fits, and when a Postgres-only tool beats a multi-database client.

Enterprise Text-to-SQL: Context, Evaluation, and Governance

A practical look at the database deployment tools DevOps teams actually use in 2026 — Flyway, Liquibase, Atlas, pgschema, Octopus Deploy, Harness, DBmaestro and Bytebase — sorted by how each one defines, applies, and coordinates schema changes.

Database Deployment Tools for DevOps Teams in 2026

A walk-through of how renaming a single column can quietly break an analytics pipeline, why it happens, and how schema change detection and notification help you catch it early.

Schema Change Detection and Notification: Why a Simple Column Rename Can Break Your Data Pipeline

How to read a Postgres query plan, spot the common performance problems, and fix them without breaking other queries.

Postgres Query Plan: How to Read and Fix Performance Issues

A walk through every layer of the Postgres permission system that can raise ERROR 42501, why each one fires, and how the behavior differs from MySQL.

The Anatomy of Postgres Permission ERROR 42501

A practical look at Postgres logs: what they record, how the logging collector works, which parameters actually matter in production, and the mistakes that turn logs into expensive noise.

Postgres Logs Explained: A Practical Guide for Production

Postgres 19 widens MultiXactOffset to 64 bits, eliminating the hidden 4 billion member limit that caused catastrophic outages for high-concurrency workloads.

Postgres 19 Feature Preview: 64-bit MultiXactOffset

How to run MySQL schema migrations without blocking queries or causing downtime, covering the MySQL-specific behaviors that bite in production and the tools that work around them.

MySQL Schema Migration Best Practice

Postgres 19 introduces automatic sequence synchronization for logical replication, eliminating manual tracking and making major version upgrades and failover significantly easier.

Postgres 19 Feature Preview: Sequence Synchronization for Logical Replication

What multi-environment database deployment actually involves, the problems teams keep hitting, and the patterns (manual scripts vs GitOps) for handling them.

Database Multi-Environment Deployments - The Challenges and Patterns

A guide to open source sensitive data discovery tools, from NLP libraries like spaCy and Microsoft Presidio to multi-source scanners like Hawk-Eye and full metadata platforms like OpenMetadata.

Top Open Source Sensitive Data Discovery Tools in 2026

Database audit logging really has two layers: the infrastructure layer your cloud provider ships, and the workflow layer you have to build yourself. Here is what each one captures, and why an audit needs both.

Database Audit Logging: Two Layers, One Trail

A practical guide to HITRUST database compliance covering access control, audit logging, change management, and data protection.

Database Compliance for HITRUST: Controls, Risks, and Best Practices

What the "Require Primary Key" SQL review rule checks, the production problems a missing primary key tends to cause, and the SQL to fix tables that don't have one.

Require Primary Key: A SQL Review Rule Explained

What the "Enforce NOT VALID in CHECK" SQL Review rule does, why adding a CHECK constraint can lock your table, and how the two-step NOT VALID approach avoids it.

Enforce NOT VALID in CHECK: A SQL Review Rule Explained

The practical ways to audit a PostgreSQL database, from native logging to triggers, logical replication, pgAudit, and Bytebase, with the trade-offs of each.

Postgres Audit Logging Guide

Learn why CASCADE operations in SQL can lead to catastrophic data loss and how the "Prohibit CASCADE" review rule protects your production database.

Prohibit CASCADE: A SQL Review Rule Explained

A walkthrough of moving each Supabase piece (database, auth, storage, functions, realtime, networking) to its AWS counterpart, with a services-first order that keeps the final database cutover simple.

How to Migrate from Supabase to AWS

Six open-source PostgreSQL backup tools compared for 2026, with the language, license, interface, and best-fit use case for each, plus a note on which ones are still maintained.

What is Enterprise Database Management (EDM)?

A plain explanation of Postgres Transaction ID (XID) wraparound: what causes it, how to check how close you are, and what keeps it from taking your database offline.

Postgres Transaction ID (XID) Wraparound

How a CI/CD pipeline for database schema migration works end-to-end. What it gates, what it leaves outside, and the tools that fit.

How to Build a CI/CD Pipeline for Database Schema Migration

A practical guide to shipping features with PostgreSQL, covering my day-to-day development workflow, the tools I actually use, and real-world examples

How I Ship Features with Postgres

What LGPD means for database operations: how it actually differs from GDPR, what the ANPD sanctions in practice, and the database controls (classification, masking, retention, audit) that hold up.

Database Compliance for LGPD: Implications and Best Practices

What a Postgres LATERAL JOIN does, when to reach for one, and the patterns and gotchas you'll run into using it.

What is Postgres LATERAL JOIN?

What MySQL Online DDL actually does, how its three algorithms (INSTANT, INPLACE, and COPY) differ, and the limitations to plan around before you run a schema change in production.

MySQL Online DDL: A Practical Guide

The best open-source Postgres connection poolers in 2026. Compare PgBouncer, Odyssey, PgDoorman, and PgDog — with our pick: PgBouncer if you want free, PgDog if you want a commercial vendor.

SQL Table Naming Dilemma: Singular vs. Plural


TypeSafe released [Jev](https://typesafe.ai/blog/introducing-system-one-models-and-jev) last week. You give it context and a list of typed questions. It returns choices, scores, or yes/no probabilities, each with a confidence number. No text generation. We wanted to see how useful that would be for database reviews.

We have used this picture for years to explain the DBA's job:

![The database operations trolley problem: a DBA at the lever, developers on one track, security engineers on the other](/content/blog/_shared/trolley-problem.webp)

Every database request is a trolley. Hold it, and the developers waiting on it get run over. Pull the lever, and the security engineers do. Jev predicted the DBA would regret the decision in 30 of our 32 cases. Sounds about right.

We tested whether Jev could help assess risk and route requests for approval. This is an experiment, with no Bytebase feature attached yet. It led us to a workflow we think makes sense:

> Database change and access workflows should combine static SQL review rules, Jev, an LLM, and human review. Static rules enforce explicit SQL policies and block violations. Jev assesses risk from the request context to determine where additional approval is needed. An LLM explains the findings and suggests fixes. A human reviews the request and makes the final call.

## The setup

We wrote eight requests a DBA might get in a week. Four are database changes:

1. A Friday 5pm hotfix: `ALTER TABLE orders ADD COLUMN ... NOT NULL DEFAULT` on a 400M-row PostgreSQL table.
1. An `UPDATE` with no `WHERE` clause, ten minutes before the billing job.
1. `CREATE INDEX` on a 2-billion-row MySQL table at peak traffic.
1. Dropping a column that "nobody uses, probably."

Four are requests for access:

1. A developer asking for read access to production `users` and `payment_methods` to debug a customer bug.
1. A full export of the `customers` table to a shared drive for a churn model.
1. `GRANT ALL PRIVILEGES` to a service account to unblock a launch.
1. An AI coding agent asking for credentials to apply its own migration to production.

We also wrote four short DBA profiles to include in the prompts:

1. **Outage veteran:** was on call for a bad outage.
1. **Ex-developer:** remembers waiting for database approvals.
1. **Compliance-first:** is three months from a SOC 2 audit.
1. **Burned-out DBA:** is the only DBA for 120 engineers.

We asked Jev about all 32 combinations of request and DBA profile. Two questions matter here:

- How bad is it for security if this runs now? Jev picks one of four levels, from negligible to a reportable event. We show them as low, moderate, and high. No request came back negligible.
- What does this DBA do in the next five minutes: pull the lever and run it now, or hold it? Jev returns a probability.

The same data drives an [interactive version of the picture](/database-trolley-problem/). Choose a DBA persona and an operation, then send the trolley. Each run samples Jev’s precomputed probabilities, so the same setup can produce different outcomes. Open “Behind the decision” to see the persona profile and model scores.

## Who picks up the ticket matters

| Type   | Request                    | Risk if it runs now | Chance the DBA runs it now, lowest to highest persona |
| ------ | -------------------------- | ------------------- | ----------------------------------------------------- |
| Change | Friday hotfix migration    | Moderate            | 6% to 93%                                             |
| Change | `UPDATE` with no `WHERE`   | High                | 1% to 41%                                             |
| Change | Index on a 2B-row table    | Low                 | 8% to 96%                                             |
| Change | Drop a column              | High                | 1% to 69%                                             |
| Access | Production read access     | Moderate            | 7% to 91%                                             |
| Access | Full customer table export | High                | 1% to 52%                                             |
| Access | `GRANT ALL` to a service   | High                | 1% to 62%                                             |
| Access | AI agent wants credentials | High                | 0% to 40%                                             |

Jev gave each request the same risk level across all four DBA profiles. Its prediction of what the DBA would do varied considerably: a 6% chance of running the Friday hotfix for the DBA burned by an outage, versus 93% for the former developer.

The DBA with the outage history was the least likely to proceed in every case. The former developer or the overloaded DBA was always the most likely. That follows the profiles we wrote: one remembers an outage, another knows the frustration of waiting for approval, and the third has 120 engineers to keep moving.

These are fictional DBAs and model predictions. Still, anyone who has worked a request queue will recognize the pattern. We want a review process that holds up regardless of who picks up the ticket. The consistent risk ratings were the result that interested us.

## Where each one fits

Here is how we would divide the review:

|                    | Bytebase SQL review               | Jev                                               | LLM                 |
| ------------------ | --------------------------------- | ------------------------------------------------- | ------------------- |
| Reads              | The SQL statement                 | The whole request: SQL, ticket, requester, timing | The whole request   |
| Returns            | Pass or fail, with the rule named | A level or a probability                          | Text                |
| Can block a change | Yes                               | No. It can add an approver                        | No                  |
| Job                | Enforce explicit SQL rules        | Assess contextual risk                            | Explain and rewrite |

**Bytebase SQL review.** The four change requests are all things [Bytebase built-in SQL review](/sql-review/) catches with a fixed rule:

- The hotfix: the _Disallow add column with default_ rule on PostgreSQL.
- The `UPDATE` with no `WHERE`: the _Require WHERE_ rule.
- The index on a 2-billion-row MySQL table: the _Limit DDL operations on tables with large data volumes_ rule.
- The dropped column: the _Backward incompatible schema change_ rule.

We would keep these checks in SQL review. The rules already catch the problem, name it, and can block the change. Asking a model adds little here.

**Jev.** The access requests need more context:

- Is `users` plus `payment_methods` a proportionate scope for debugging one customer's bug?
- Should a full customer table land in a shared drive because a board deck is due?
- Does a service that fails on one permission error need `ALL PRIVILEGES`?
- Should an agent hold production write credentials because its squad has a demo tomorrow?

To judge these requests, a reviewer needs the ticket, the requester, the timing, and what the tables contain. Jev can take that context and return a risk level for the approval workflow to use.

**LLM.** The developer whose hotfix is held needs to know why and how to split it into an add, a backfill, and a constraint. The approver wants a three-line summary of a 300-line migration. This is where we would use an LLM.

**Human approver.** The approver gets the rule violations, the risk rating, and the explanation, then decides whether the request can proceed.

## Why try Jev for risk ratings?

Teams already ask LLMs for risk ratings in JSON. Jev is worth trying for that job for a few reasons:

1. **Typed output.** Jev returns the requested value directly, without asking a text model to follow a JSON format.
2. **Consistent ratings in this test.** Changing the DBA profile left each request's risk level unchanged. We would need broader testing before relying on that in an approval policy.
3. **A confidence number.** This could help route uncertain cases for closer review, provided the confidence is well calibrated. TypeSafe says LLMs tend to be overconfident when asked to rate their own answers; we did not test that comparison.
4. **Low cost.** Our 32 calls cost less than a cent. TypeSafe claims 70 to 500ms per call, which would be practical for a review workflow.

## How it would fit the workflow

Bytebase routes approvals by condition: environment, project, engine, SQL type, affected rows, and a risk level of high, moderate, or low. Requests for a role and for just-in-time access go through the same approval flows as schema changes.

We mapped Jev's ratings to those three levels so they could feed into the existing routing. A high rating could add an approver to a change that passes SQL review, or flag an access request for closer review.

We would let the rating add scrutiny, with every policy-required approval still in place. If production access needs a DBA's sign-off, a low score from Jev should leave that requirement alone.

## What we did not test

- **Intent.** We did not plant a mismatch between the SQL and the ticket in any of the eight requests, so we learned nothing about whether Jev could catch one.
- **Calibration.** We had no ground-truth labels to check the ratings or confidence numbers against. The agreement across DBA profiles tells us little about accuracy.

## References

- [Introducing System One Models & Jev](https://typesafe.ai/blog/introducing-system-one-models-and-jev)
- [SQL review rules](https://docs.bytebase.com/sql-review/review-rules)


The
Bytebase Blog

How to Prevent an AI Agent From Dropping Your Production Database

What Postgres is Missing for AI Agents

Explore the standard for database governance

Solutions & Platform

Resources

Change Management

Access & Security

AI & Agents

Comparisons

Company

Legal

The Bytebase Blog

How to Prevent an AI Agent From Dropping Your Production Database

What Postgres is Missing for AI Agents

Explore the standard for database governance

The
Bytebase Blog