# LATEST ## Why We Refreshed Greptile's Brand We gave Greptile a brand refresh, entirely in-house. Not just a visual update—a way to give a technical product a clearer shape. ## Slop Is Not Necessarily The Future Everyone's worried about slop, but good code will prevail, not only because we want it to, but because economic forces demand it. ## Hands-On Evaluation of NVIDIA Nemotron 3 Super We tested NVIDIA Nemotron 3 Super 120B on real PRs using our internal eval harness. It punches far above its weight class. ## Greptile v4 + New Pricing Announcing Greptile Agent v4, our best code review agent yet. 74% more addressed comments and 43% comment acceptance rate. ## There is an AI Code Review Bubble Today everybody's doing AI code review. Here's how Greptile's viewpoint is differentiated - independence, autonomy, and feedback loops. ## Every GitHub Object Has Two IDs How I reverse-engineered GitHub node IDs to avoid a database migration and discovered their dual ID system along the way. ## Greptile v3, an agentic approach to code review Announcing Greptile v3, a complete rewrite of our code review workflow that uses an agentic approach to autonomously validate code with 256% better upvote/downvote ratios and 70.5% higher acceptance rates. ## Sandboxing agents at the kernel level Tracing the open syscall to understand how containers conceal files from agents. ## Series A and Greptile v3 Announcing Greptile's $25M Series A led by Benchmark Capital and unveiling Greptile v3—our most advanced AI code reviewer that catches 3x more critical bugs. ## Greptile's work culture A from-the-source outline of our work culture, so potential teammates can self-assess their fit. ## Software Needs An Independent Auditor Explore why Greptile separates code generation from code review and the importance of independent auditing for AI-generated code. ## What Developers Need to Know About AI Code Reviews What we learned from reviewing 700K+ pull requests per month: why AI-generated code needs an independent reviewer, how confidence scores change PR triage, and what real catches look like. ## Greptile's Biggest Update Yet Greptile’s BIGGEST update yet! Long-term memory, MCP integrations, scoped rules & 40% cheaper pricing - launching today! ## Do Larger PRs Get Merged Faster? 24,000+ PRs analyzed reveal larger PRs merge FASTER + team size impacts nobody expects. ## AI Code Review: Should the Author Be The Reviewer? AI reviewing AI code? Exploring the paradox of using AI to review AI-generated code & whether this creates conflict or genius. ## Codebases are uniquely hard to search semantically Explore why semantic search in codebases is challenging and how natural language processing and granular chunking improve AI code understanding. ## Is Two Reviewers the New Standard? Exploring whether two reviewers per PR is the sweet spot for code quality, team velocity, and collaboration. ## The Content-ification of Software Explore how AI will transform software development by enabling highly specialized software companies to profitably service extremely niche markets. ## How to Make LLMs Shut Up Lessons learned from our challenges reducing LLM output: Discover everything that went wrong when trying to make our models say less. ## Splitting engineering teams into defense and offense Discover how Greptile, an AI-powered code review startup, balances event-driven and long-running engineering tasks to maximize productivity in a small team. ## Introducing Greptile 2.0 Discover Greptile 2.0, the AI platform that understands your codebase—boost development with smart code reviews, JIRA integration, and custom workflows. ## Announcing our $4.1M seed round Greptile raises $4.1M seed round to transform AI code review with their innovative API. Learn how they're helping 400+ teams optimize development workflows.