Qodo 2.1 adds AI-driven rules for smarter code review

Qodo has released Qodo 2.1, introducing a rules management layer for AI-assisted code review. The update replaces static rule files with a system that learns from code and pull request decisions over time.

Many software teams keep coding standards in markdown documents, internal wikis, prompts, or tool-specific files such as .cursorrules. Those standards can drift as teams change and codebases grow. AI coding tools may also treat the guidance as optional, leaving engineering leaders with limited visibility into adherence and outcomes.

Qodo positions the release as a response to this governance gap as AI-generated code volumes rise. It describes the Rules System as a "continuous learning" approach that creates a central source of truth for coding standards and applies them consistently during pull request review.

Rules Layer

The system adds automated rule discovery, ongoing maintenance, enforcement during pull request review, and analytics. Qodo says it derives rules from code patterns and historical pull request decisions rather than requiring engineers to write and maintain rule text by hand.

It also flags issues that can erode rule sets over time, including conflicts, duplicates, and out-of-date standards. This targets a common problem in large engineering organisations, where different teams introduce overlapping guidance across tools and documents.

Enforcement is built into the code review workflow. Rules are applied programmatically during pull request review, and the product recommends fixes when it finds violations. Analytics track adoption and violation trends, with reporting intended to show whether standards lead to measurable improvements.

Overall, the release shifts governance from a collection of documents and prompts to a managed system that can be audited and measured. The framing aligns with broader changes in software development, where teams increasingly treat policy and controls as code and rely on automated feedback loops.

Visibility Push

Qodo argues that static files do not provide enough visibility into what an AI coding assistant followed or whether standards improved outcomes. Without measurement, it says, engineering teams struggle to determine which rules matter and which create noise in the review process.

"Engineering standards shouldn't be scattered across docs, linters, and engineer's heads," said Itamar Friedman, CEO and Co-Founder of Qodo. "AI is producing more code than ever, which means organizations need their standards to be just as intelligent as their AI coding tools. Qodo's Rules System enables organizations to define standards based on real behavior, maintain them automatically, and prove their impact with real analytics. This is how you scale code quality in the age of AI."

Customer Example

Hibob says it has been using the Rules System to consolidate standards spread across repositories and other references. It links the approach to improved consistency across teams and faster onboarding-common goals for engineering organisations operating across multiple products and services.

"Maintaining consistent standards is critical for our engineering teams," said Ofer Morag Brin of Hibob. "Qodo's Rules System didn't just surface the standards we had scattered across different places; it operationalized them. The system continuously reinforces how our teams actually review and write code, and we are seeing stronger consistency, faster onboarding, and measurable improvements in review quality across teams."

Product Context

Qodo sells an AI code review and quality product aimed at enterprise engineering teams. Alongside the rules layer, Qodo 2.1 also includes what it calls "advanced context engineering" and a multi-agent review system. Qodo says the approach draws on full-repository signals, including codebase history and prior pull request decisions, with an emphasis on explainable feedback and reduced noise.

The Rules System can be centrally managed through a single portal and applied across an organisation's development toolchain, with analytics reporting on adoption, violations, and improvement trends.

Founded in 2018, Qodo has raised $50 million. Backers include TLV Partners, Vine Ventures, Susa Ventures, and Square Peg, along with angel investors including executives from OpenAI, Shopify, and Snyk.