{"slug": "one-github-issue-six-perspectives-how-ai-review-profiles-improve-software", "title": "One GitHub Issue, Six Perspectives: How AI Review Profiles Improve Software Estimation", "summary": "A developer built Pointwise, a GitHub-integrated Planning Poker app that uses GitHub Copilot to review issues through six optional professional review profiles—Engineer, UX Designer, Product Owner, QA Engineer, Business Analyst and Security Engineer—to surface hidden assumptions and edge cases during backlog refinement. The tool aims to counter the tendency to estimate requirements only through one's own professional lens, which can lead to underestimated scope once development begins.", "body_md": "Most developers have experienced this during backlog refinement.\n\nA GitHub issue looks straightforward. The requirements seem reasonable, the implementation appears manageable, and the team agrees on an estimate.\n\nThen development starts.\n\nSomeone identifies a missing error state. QA raises an edge case. A security concern emerges. The Product Owner clarifies a requirement that changes the scope entirely.\n\nSuddenly, that five-point issue looks considerably more complicated.\n\nThe problem isn't necessarily poor estimation. It's that **we tend to evaluate requirements through the lens of our own experience.**\n\nWhat happens when we deliberately look at the same issue from different professional perspectives?\n\nImagine a GitHub issue with the following requirement:\n\nAllow users to update their notification preferences, including email notifications and in-app alerts.\n\nAs a developer, my immediate questions might be:\n\nThese are sensible technical questions.\n\nBut they aren't the only questions worth asking.\n\nA UX Designer might focus on how users understand the available options.\n\nA QA Engineer might consider what happens when a user changes preferences across multiple devices.\n\nA Security Engineer might question whether users can modify another account's preferences.\n\nAll of these perspectives can influence the implementation, testing effort and ultimately the estimate.\n\nThis is the thinking behind the optional [AI review profiles in Pointwise](https://pointwise-estimates.tech/), a GitHub-integrated Planning Poker application.\n\nPointwise uses GitHub Copilot to review an issue from a selected professional perspective.\n\nRather than asking AI to provide a generic assessment, contributors can choose a profile that influences which aspects of the requirements receive attention.\n\nThere are currently six profiles.\n\nThe Engineer profile focuses on technical implementation.\n\nFor our notification preferences example, useful questions might include:\n\nThe objective is to expose technical assumptions and potential implementation complexity.\n\nThis is particularly useful when a requirement appears simple at the UI level but may involve several backend services.\n\nDevelopers often receive requirements describing what a feature should do without fully explaining how users will interact with it.\n\nThe UX Designer perspective encourages questions about usability, accessibility and interaction behaviour.\n\nFor example:\n\nThese aren't merely visual design considerations.\n\nThey can introduce additional states, components and acceptance criteria that affect development effort.\n\nA technically complete implementation isn't necessarily a useful product feature.\n\nThe Product Owner profile considers requirements in terms of user value, scope and intended outcomes.\n\nFor our example:\n\nThis perspective can reveal requirements that are either missing or unnecessarily broad.\n\nSometimes the most valuable discovery is that the team can deliver a smaller, more focused solution.\n\nQA engineers are particularly good at identifying situations that aren't covered by the happy path.\n\nConsider the notification preferences feature.\n\nWhat happens if the API request fails?\n\nWhat if the user submits changes twice?\n\nWhat if preferences are modified from another device?\n\nWhat happens when an existing user has no saved preferences?\n\nThe QA Engineer perspective encourages examination of edge cases, validation, error handling and testability.\n\nThese questions matter because unexpected behaviour doesn't become cheaper to address simply because nobody mentioned it during refinement.\n\nThe Business Analyst profile looks at requirements, business rules and dependencies.\n\nIt can help identify ambiguity between what stakeholders expect and what developers believe they're implementing.\n\nThis perspective is particularly useful for enterprise applications where seemingly isolated features often interact with wider business processes.\n\nSecurity requirements can be overlooked when the discussion focuses primarily on delivering functionality.\n\nFor notification preferences, a security-focused review might ask:\n\nNot every issue needs a lengthy security assessment.\n\nBut identifying an authorisation problem before development is considerably better than discovering it after release.\n\nThere's an important distinction between using AI to review requirements and allowing AI to determine how much work something involves.\n\nAI doesn't necessarily understand the complete codebase, the team's experience or every architectural constraint.\n\nIt can also misunderstand requirements or raise concerns that aren't relevant.\n\nThat's why I see these reviews as **additional perspectives rather than authoritative answers.**\n\nIf an AI review suggests that a database migration might be required, the team should verify whether that's true.\n\nIf it identifies an accessibility concern, developers should evaluate the actual interaction.\n\nThe value comes from prompting useful conversations, not blindly accepting generated recommendations.\n\nIn [Pointwise's collaborative Planning Poker workflow](https://pointwise-estimates.tech/), contributors estimate independently before their scores are revealed.\n\nThe team can then discuss differences and reach consensus.\n\nAn optional Copilot review provides another way to examine the issue, particularly when requirements are unclear or contributors have identified competing assumptions.\n\nFor example, a team might initially estimate five points.\n\nA QA-focused review could raise several missing error states.\n\nAfter investigating those concerns, the team might decide the original estimate remains appropriate, or that the scope needs clarification.\n\nEither outcome is useful.\n\nThe AI hasn't replaced the team's judgement. It's helped the team ask better questions.\n\nI've built these review profiles into [Pointwise](https://pointwise-estimates.tech/) to make it easier to explore requirements from different perspectives without leaving the GitHub issue and estimation workflow.\n\nThe walkthrough below demonstrates the collaborative estimation process and optional Copilot review.\n\n[Watch the Pointwise walkthrough on YouTube](https://youtu.be/AyxBZgLLidA?si=lAzb_COrcrbwNb_S)\n\nSoftware estimation isn't just about understanding how difficult something will be to code.\n\nIt's about understanding the work well enough to make an informed decision.\n\nEngineers, designers, testers, product owners, business analysts and security specialists all approach requirements differently.\n\nThose differences are valuable.\n\nAI review profiles offer a practical way to introduce additional questions into refinement, especially when every specialist isn't available for every discussion.\n\nBut the responsibility for evaluating those questions must remain with the team.\n\n**The goal isn't to let AI tell developers what they've missed. It's to help developers discover what might be worth discussing before implementation begins.**\n\n**Which perspective tends to uncover the most unexpected complexity in your team's requirements: engineering, UX, QA, product or security?**", "url": "https://wpnews.pro/news/one-github-issue-six-perspectives-how-ai-review-profiles-improve-software", "canonical_source": "https://dev.to/ben_segni_4c5eacf9af147ab/one-github-issue-six-perspectives-how-ai-review-profiles-improve-software-estimation-2i9f", "published_at": "2026-11-11 11:38:34+00:00", "updated_at": "2026-11-11 11:52:01.657927+00:00", "lang": "en", "topics": ["ai-tools", "ai-products", "developer-tools", "ai-agents"], "entities": ["Pointwise", "GitHub", "GitHub Copilot"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/one-github-issue-six-perspectives-how-ai-review-profiles-improve-software", "markdown": "https://wpnews.pro/news/one-github-issue-six-perspectives-how-ai-review-profiles-improve-software.md", "text": "https://wpnews.pro/news/one-github-issue-six-perspectives-how-ai-review-profiles-improve-software.txt", "jsonld": "https://wpnews.pro/news/one-github-issue-six-perspectives-how-ai-review-profiles-improve-software.jsonld"}}