Pull request
TrialTools
A code review workflow for proposing, discussing, and merging changes into a codebase.
Why it's here
Placed in Trial: 28 article(s) of evidence from 5 source(s), led by research-stage coverage, with 5 in the last 30 days. Confidence 94%.
Evidence (19)
- 6The New Stack·9/18/2026researchAI code review is overloading senior engineers
The article argues that heavy AI-assisted coding is shifting senior engineers from writing code to reviewing large volumes of generated pull requests, creating review bottlenecks and burnout. It says AI output is often plausible but flawed, and recommends codifying recurring review feedback into rules and tests to reduce manual review load.
- 7The New Stack·9/17/2026researchHuman oversight shifts from code to requirements in AI development
The article argues that AI-assisted software pipelines now have strong controls for checking whether code matches specifications, but weak controls for validating the requirements themselves. It describes a workflow where recorded meetings, scoping documents, unit specs, human reviews, and automated checks can still certify a system built from a flawed requirement, highlighting the need for human oversight earlier in the process.
- 5InfoQ·9/16/2026product_launchDuolingo Uses AI Training to Automate Code Review
Sarah Deitke describes how Duolingo paired internal AI literacy workshops with observability dashboards to encourage broader AI adoption across engineering teams. She also presents a case study where an automated pull request risk-assessment bot was used to redesign code review while maintaining defect rates and speeding delivery.
- 8The New Stack·9/9/2026securityOpenAI lets AI block engineer code merges
OpenAI has made automated security review mandatory for engineer pull requests, and its AI models can prevent a merge if they detect a vulnerability. The company also uses models for code review, regression detection, dependency upgrades, and maintenance tasks, with some internal benchmarks described as “superhuman” in code review.
- 4InfoQ·9/3/2026researchSustainable Human Code Review in the AI Era
The article discusses a review process that combines mandatory AI checks with selective manual review spikes for complex changes. It also suggests skipping peer review on low-risk pull requests and using AI approvals in small teams where most developers are code owners.
- 7InfoQ·8/22/2026researchLinkedIn Builds Multi-Agent AI Code Review System
LinkedIn engineers created a multi-agent AI code review platform to handle pull requests at company scale. The system is designed around LinkedIn's internal coding context and aims to reduce hallucinations and low-signal feedback compared with generic off-the-shelf reviewers.
- 2Hacker News·8/14/2026researchHN post criticizes oversized AI-generated pull requests
A Hacker News post argues that AI-assisted development is leading to oversized pull requests that are harder for humans to review and maintain. The author says small PRs are meant to improve reviewability and asks developers to avoid shipping large, hard-to-digest diffs, even when AI can generate them quickly.
- 6The New Stack·8/13/2026framework_updateCode review shifts from diff checking to product judgment
The article argues that code review is evolving beyond line-by-line bug catching into a broader decision-making process centered on judgment, collaboration, alignment, and verification. As AI can generate large amounts of code quickly, teams may need to review intent and product decisions earlier rather than only inspecting the final diff.
- 6The New Stack·8/12/2026product_launchCodeRabbit launches Agentic Change Management for AI-driven software change
CodeRabbit announced an Agentic Change Management control layer to help teams govern software changes created by both human developers and AI agents. The company argues that issue tracking is becoming less central as pull requests turn into the main auditable decision point for planning, review, and shipping code.
- 7The New Stack·8/9/2026researchAI speeds up coding, but engineering output stays flat
DX’s State of AI Impact in Engineering report says AI tooling has made individual developers faster, but organizational velocity has not improved in step. The report points to growing coordination overhead, flat innovation ratios, and weaker change confidence, especially in larger companies and larger pull requests.
- 4The New Stack·8/8/2026researchAI Adoption vs. AI Usage in Engineering Teams
The article argues that common AI adoption metrics, such as seat activations, token spend, and weekly active usage, mostly measure usage rather than real workflow change. It says lasting adoption only happens when AI becomes part of team norms or recurring processes, not just individual experimentation.
- 6InfoQ·8/7/2026product_launchSpotify Details AI Agent for Fleet-Wide Codebase Migrations
Spotify engineers Jo Kelly-Fenton and Aleksandar Mitic describe Honk, an AI coding agent built to manage large-scale codebase migrations across thousands of repositories. The talk focuses on separating CI verification runtime from the agent, handling pull request bottlenecks, and enforcing standardization across the engineering fleet.
- 4InfoQ·8/7/2026framework_updateRootly Ends Small PR Rule as AI Agent Code Generation Grows
Rootly says its long-standing rule favoring small pull requests is no longer effective because AI agents now generate most of its code. The company is shifting review criteria toward blast radius, using feature flags and rollback readiness to judge risk instead of line count.
- 6Hacker News·8/5/2026open_sourceRust Project Adopts Public Policy for LLM Use in rust-lang/rust
Several teams in the Rust project have adopted a new policy governing how large language models may be used in contributions to the rust-lang/rust monorepo. The policy is intended to clarify expectations for contributors and reviewers, especially around LLM-generated code, issue reporting, and quoted LLM output.
- 8InfoQ·7/28/2026securityGitHub hardens npm and Actions against supply chain attacks
GitHub says it has rolled out several changes to disrupt common supply chain attack techniques targeting npm and GitHub Actions. The updates include temporary protection for high-impact npm accounts, safer defaults for actions/checkout, workflow trigger controls, and read-only cache access for untrusted Actions runs.
- 6The New Stack·7/23/2026researchHow regulated organizations can safely speed up AI coding
The article argues that regulated organizations can use AI to increase software delivery speed, but only if verification becomes a continuous part of the engineering process. It highlights how AI-assisted development and agentic workflows can help domain experts and engineers collaborate more effectively while maintaining accountability, security, and compliance.
- 5The New Stack·7/19/2026researchAI Is Pushing Code Review Upstream
The article argues that traditional pull-request-based code review is becoming a poor fit for AI-assisted software development, where large amounts of code are generated quickly. It suggests teams should review intent, specs, and acceptance criteria earlier in the workflow rather than waiting until merge time.
- 4The New Stack·7/16/2026researchAI Hasn't Moved the Software Bottleneck to Code Review
The article argues that AI coding tools have not shifted the main bottleneck from coding to code review. Instead, it says the real constraint is usually downstream in testing, deployment, and release batching, where changes pile up before reaching users.
- 6The New Stack·7/11/2026researchAI code reviews expose a new software bottleneck
The article argues that merge quality should be treated as an explicit contract, especially as coding agents increase pull request volume. It says traditional pipelines usually verify only the first three layers of confidence, while real system behavior testing remains the costly missing layer that often catches microservice bugs.