Bemi

Bemi is an audit trail tool for PostgreSQL databases that tracks and records data changes automatically, without requiring custom engineering work or extra infrastructure. It connects to an existing PostgreSQL database and captures every change, including ones made through direct SQL outside the application. Use cases include audit and compliance, observability and troubleshooting, data recovery, and building an activity feed. It is used by engineering teams that need a reliable source of truth for database changes.

Bemi integrates at the database level by connecting to PostgreSQL's Write-Ahead Log and using Change Data Capture to identify and capture data changes in real time, without requiring write permissions or changes to table structure. At the application level, open-source libraries for popular ORMs add context such as the user, API endpoint, or worker behind each change. Bemi workers combine the low-level data with this application context and store it in a structured, queryable format inside a PostgreSQL database.

Bemi tracks changes with full reliability, even when they are made through direct SQL outside the application, without affecting code runtime or database workload. It supports time travel querying so historical data can be retrieved, grouped or filtered by application context, and it keeps an audit trail for compliance. Data is stored in an auto-scaled, optimized serverless PostgreSQL database that the customer owns, with encryption of data at rest and in transit and customer-level infrastructure isolation.

Bemi offers usage-based plans with a 14-day free trial, standard support and a dedicated Slack channel on every plan. The Starter plan costs $19/month for 1K new data changes per month, unlimited retention, 1 connected database and 3 dashboard users. The Business plan starts from $99/month for 50K+ data changes per month and up to 3 connected databases. An Enterprise plan with custom data volume and unlimited databases is available as a cloud or self-hosted deployment, with pricing by contacting Bemi.

Use cases

• Audit and compliance logs surfaced to customer support and external customers
• Reverting changes made by a user or rolling back an API request
• Retrieving historical data without implementing event sourcing
• Troubleshooting application issues by identifying their root cause
• Tracking changes across distributed systems
• Analyzing historical trends for informed decision-making