The author of a blog post on Greptile, an AI code review tool, reflects on the current boom in the AI code review space. With competitors like OpenAI, Anthropic, and others entering the market, the post questions how to differentiate a product. The author argues that performance is subjective. Instead, they focus on their viewpoint for future code review.

The core of Greptile's strategy centers on three pillars: independence, autonomy, and feedback loops. They believe the review agent should be separate from the coding agent. This contrasts with some tools that combine coding and review. They envision a future where agents handle a large percentage of code approval, reducing human involvement in routine checks.

Greptile aims for full automation of code validation through review, testing, and QA. This is work that humans often dislike. Also, they've built a Claude Code plugin to address comments and run a feedback loop. The goal is a system where humans express intent, agents code, and review tools provide feedback.

The proliferation of AI code review tools has accelerated recently. This has forced companies to think about long-term strategy. Greptile's approach emphasizes a future where AI handles more of the code review process. It remains to be seen how the market will shake out as more companies adopt these tools.