CNCF TLV × AI Talks IL: Rethinking Code Reviews & Agents in the AI Era

Code review is treated as the final safety net of modern software engineering. That assumption is quietly breaking down as AI writes most of our code.

In many teams today, LLMs generate refactors, tests, fixes, and migrations. The author reviews the code, runs tests, validates behavior, and understands the change end to end. Then they are still required to perform the sacred ritual of opening a pull request and waiting for approval. Days later, an async human review arrives from someone with less context and worse tooling. The result looks like rigor, but mostly adds delay and shallow feedback.

This talk argues that traditional PR review has become theater in an AI-driven workflow. AI tools already catch style issues, missing error handling, security risks, and best-practice violations faster and more consistently than humans. Meanwhile, the failures that actually matter, performance, edge cases, operational impact, are rarely caught by late, asynchronous reviews anyway.

Based on real production workflows, the session proposes a shift to AI-assisted self-review - engineers own their changes end-to-end, AI enforces consistency and standards, and human review moves to where it adds real value - design discussions, architecture decisions, and mentorship.

If AI has fundamentally changed how we write code, it’s time to change how we decide it’s safe to ship.

Most AI code review implementations focus on the wrong metrics—counting comments generated or code accepted rather than measuring developer velocity and code quality improvements. The real value lies in intelligent context integration and organizational learning at scale. This talk examines successful AI code review deployments across engineering organizations, revealing four critical success factors: seamless integration with your existing development context (codebase, tickets, architectural decisions), specialized review guidelines that scale from 5 to 50,000 repositories, comprehensive observability to understand where AI adds value versus where it creates noise, and strategic human-AI collaboration patterns. We'll explore real case studies of teams who've moved beyond basic linting to utilizing AI code review that understands business logic, catches architectural anti-patterns, and actually accelerate PR cycles. You'll learn when AI reviews shine, when they don't, and how to measure the difference. This isn't about replacing human reviewers, it's about building an intelligent system that amplifies human expertise and reduces cognitive load where it matters most.

AI agents are moving beyond interactive chatbots into our core infrastructure, becoming active components in CI/CD pipelines, automated remediation, and DevOps workflows. Operating agents headlessly within continuous delivery systems introduces distinct engineering challenges: configuring non-interactive runs, bundling dynamic Agentic Features into versioned artifacts, and managing operational reliability in production. In this talk, we will explore the concept of agents as versioned, runnable packages. We’ll discuss how to treat agentic workflows with the same rigors as traditional application code, touching on containerization as a packaging strategy without treating containers as the sole focus, and review the tooling and patterns needed to reliably ship scriptable agents into modern DevOps pipelines.