--- title: How to Perform Effective Project Management with AI description: An engineer shares personal workflow tips for managing software projects when most coding is delegated to AI agents. The suggested time split shifts from... canonical: https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How to Perform Effective Project Management with AI | daily.dev og:description: An engineer shares personal workflow tips for managing software projects when most coding is delegated to AI agents. The suggested time split shifts from... og:url: https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc og:image: https://api.daily.dev/og/posts/JKp7tKjwC.png og:image:alt: How to Perform Effective Project Management with AI og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # How to Perform Effective Project Management with AI **[Towards Data Science](https://daily.dev/sources/tds)** · 8 min read · 1 upvotes · 0 comments ## Summary An engineer shares personal workflow tips for managing software projects when most coding is delegated to AI agents. The suggested time split shifts from mostly writing code to prompting agents, testing, and meetings, with extra time freed up for other work. Techniques include mapping out task ambiguities before handing work to agents, using a custom /goal command to force agents to keep working until tasks are truly complete, and wiring up Playwright MCP so agents can test their own changes end-to-end in a browser. The piece also notes anecdotal laziness differences between model versions referred to as Opus 5, Opus 4.8, and Fable 5. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I get Claude Code or Codex agents to run longer without stopping to ask me questions? Map out the task thoroughly beforehand so ambiguities are resolved before the agent starts, rather than clarifying mid-run. One approach has the agent present all ambiguities in an HTML report for the developer to answer up front, after which the agent can work autonomously for extended periods until the code reaches a dev branch. _Developers refining autonomous agent workflows compare tactics like this through daily.dev._ ### How can I stop an AI coding agent from stopping work before a task is actually finished? Use a hook, referred to as a /goal command, that triggers whenever the agent claims completion and forces it to reflect on whether all requested work was actually done, prompting it to keep going if not. This is described as useful because models like Opus 5 can otherwise behave lazily and stop short of full completion. _Anyone tuning agent persistence and completion behavior can track such workarounds via daily.dev._ ### How can I automate end-to-end testing for code written by AI agents? Give agents access to Playwright MCP so Claude Code and Codex agents can spin up localhost servers, open Chrome, and click through the application themselves to verify behavior, such as catching a button that leads to a 404 page. This offloads manual verification and addresses testing becoming the new time bottleneck once coding itself is largely automated. _Teams shifting testing effort onto agents can follow practical setups like this on daily.dev._ ## Similar posts on daily.dev - [How Humans and AI Agents Can Work Together: A Practical Guide to Agent-Based Project Management](https://daily.dev/posts/how-humans-and-ai-agents-can-work-together-a-practical-guide-to-agent-based-project-management-lllaxvzfz) · DEV · 1 upvotes · 0 comments - [How to Organize All of Your Coding Agent Tasks](https://daily.dev/posts/how-to-organize-all-of-your-coding-agent-tasks-ts8v4485a) · Towards Data Science · 2 upvotes · 0 comments - [How to get ahead of 99% of software engineers with AI agents](https://daily.dev/posts/how-to-get-ahead-of-99-of-software-engineers-with-ai-agents-n9pug1pul) · System Design Newsletter · 24 upvotes · 1 comments - [My Thoughts on AI, Part 2: Agent Setup, Workflow, and Tools](https://daily.dev/posts/my-thoughts-on-ai-part-2-agent-setup-workflow-and-tools-bwcpsofep) · Mark Erikson · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#claude-code](https://daily.dev/tags/claude-code), [#project-management](https://daily.dev/tags/project-management), [#vibe-coding](https://daily.dev/tags/vibe-coding) [View this post on daily.dev](https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"How to Perform Effective Project Management with AI","url":"https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc"},"datePublished":"2026-08-17T13:53:45.221Z","dateModified":"2026-09-14T07:31:06.386Z","description":"An engineer shares personal workflow tips for managing software projects when most coding is delegated to AI agents. The suggested time split shifts from...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/1dbd554b768052aeb4688f5fa22e8d45?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/1dbd554b768052aeb4688f5fa22e8d45?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Towards Data Science","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Towards Data Science","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tds","url":"https://daily.dev/sources/tds"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,claude-code,project-management,vibe-coding","timeRequired":"PT8M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Towards Data Science","item":"https://daily.dev/sources/tds"},{"@type":"ListItem","position":3,"name":"How to Perform Effective Project Management with AI"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc#faq","mainEntity":[{"@type":"Question","name":"How can I get Claude Code or Codex agents to run longer without stopping to ask me questions?","acceptedAnswer":{"@type":"Answer","text":"Map out the task thoroughly beforehand so ambiguities are resolved before the agent starts, rather than clarifying mid-run. One approach has the agent present all ambiguities in an HTML report for the developer to answer up front, after which the agent can work autonomously for extended periods until the code reaches a dev branch. Developers refining autonomous agent workflows compare tactics like this through daily.dev."}},{"@type":"Question","name":"How can I stop an AI coding agent from stopping work before a task is actually finished?","acceptedAnswer":{"@type":"Answer","text":"Use a hook, referred to as a /goal command, that triggers whenever the agent claims completion and forces it to reflect on whether all requested work was actually done, prompting it to keep going if not. This is described as useful because models like Opus 5 can otherwise behave lazily and stop short of full completion. Anyone tuning agent persistence and completion behavior can track such workarounds via daily.dev."}},{"@type":"Question","name":"How can I automate end-to-end testing for code written by AI agents?","acceptedAnswer":{"@type":"Answer","text":"Give agents access to Playwright MCP so Claude Code and Codex agents can spin up localhost servers, open Chrome, and click through the application themselves to verify behavior, such as catching a button that leads to a 404 page. This offloads manual verification and addresses testing becoming the new time bottleneck once coding itself is largely automated. Teams shifting testing effort onto agents can follow practical setups like this on daily.dev."}}]} ```