GitHub Stacked Pull Requests Enter Public Preview, Promising Faster Code Reviews

·1 views

GitHub launched stacked pull requests into public preview on July 30, 2026, allowing developers to break large features into an ordered series of smaller, focused pull requests. In this workflow, each PR targets the one below it rather than the main branch, so reviewers see only the changes relevant to a single layer — such as data models, API logic, state management, or UI. The approach addresses a common bottleneck where monolithic pull requests spanning thousands of lines either receive superficial sign-offs or cause lengthy review cycles with cascading rework. Previously, developers who manually chained feature branches faced painful rebasing whenever an early layer needed changes, often abandoning the strategy altogether. Stacked PRs aim to make incremental, layer-by-layer review practical without the manual overhead.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Automation Bias Makes AI Human-in-the-Loop Oversight Largely Ineffective

Automation bias is the tendency to over-trust automated systems, leading people to approve AI-generated outputs without meaningful scrutiny or to stop monitoring them altogether. Research on AI coding agents found that even when humans were required to review and approve an agent's plan before it acted, they still failed to catch harmful actions roughly three times out of four. The failure was not simple inattention — reviewers saw the actions but rationalized them, a phenomenon researchers described as a 'recognition bottleneck' where noticing something does not translate into identifying it as a problem. This means adding an approval step to an AI workflow can create a false sense of safety, with audit logs recording human sign-off on decisions that were never genuinely evaluated. Experts argue the problem is structural rather than a character flaw, worsening as agents become more reliable, and that effective oversight requires deliberate interface and process design rather than just inserting a human checkpoint.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Ashby's isRemote field flags hybrid jobs as remote, skewing job board data

A developer analyzed 1,668 job postings from ten company boards using Ashby's public job board API on August 2, 2026, and found that the isRemote boolean field returns true for both remote and hybrid roles, not just fully remote ones. Cross-tabulating isRemote against the workplaceType field across 1,637 postings confirmed that isRemote is effectively equivalent to workplaceType not being 'OnSite', meaning hybrid roles are included. This causes significant overstatement of remote openings — OpenAI's board, for example, showed 485 isRemote postings but only 32 genuinely remote roles, a 15x inflation. The author recommends using the three-value workplaceType field (Remote, Hybrid, OnSite) instead of the boolean, and treating missing workplaceType as 'Unknown' rather than false. A separate finding noted that compensation data is omitted from API responses unless the includeCompensation=true parameter is explicitly added to requests.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

No Evidence OpenAI Has Launched a Formal 'Team Humanity' Initiative

The phrase 'Team Humanity' has circulated in connection with OpenAI, raising speculation about a possible governance or safety program, but no official announcement or documentation supports its existence. Searches of OpenAI's public materials as of August 1, 2026 found no program, team, or policy initiative carrying that name. The term also appears in unrelated humanitarian contexts, increasing the risk of mistaken attribution to OpenAI. Analysts caution that businesses should not adjust technical roadmaps or compliance plans based on an unverified label lacking defined scope or leadership. Should OpenAI confirm such an initiative in the future, developers and enterprise customers would need specifics on policy changes, API terms, and governance accountability before taking action.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

AI Tools Amplify Both Skill and Incompetence, Warn Engineering Experts

A widely discussed analysis argues that large language models in software development function like power tools — accelerating output for skilled developers while magnifying errors made by inexperienced ones. Experienced engineers can use LLMs to explore codebases, generate tests, and compare designs faster, but the tools cannot substitute for deep technical judgment about system design and architecture. The piece cautions that beginners risk skipping foundational learning — such as debugging, data modeling, and understanding control flow — if they rely on AI-generated code before developing the mental models needed to evaluate it. Beyond individual skill, the article contends that scaling software quality requires standardized processes, defined components, and systemic quality controls, much like IKEA's manufacturing model, rather than simply equipping every developer with more powerful tools.

0 comments Read more at DEV Community