--- title: The AI-native SDLC won’t be one process | daily.dev description: Anthropic's AI-Native SDLC Playbook argues that with AI agents, code generation is no longer the bottleneck and process (planning, review, verification,... canonical: https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: The AI-native SDLC won’t be one process | daily.dev og:description: Anthropic's AI-Native SDLC Playbook argues that with AI agents, code generation is no longer the bottleneck and process (planning, review, verification,... og:url: https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x og:image: https://api.daily.dev/og/posts/ecJTxEG8X.png og:image:alt: The AI-native SDLC won’t be one process 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. # The AI-native SDLC won’t be one process **[The New Stack](https://daily.dev/sources/newstack)** · 7 min read · 1 upvotes · 0 comments ## Summary Anthropic's AI-Native SDLC Playbook argues that with AI agents, code generation is no longer the bottleneck and process (planning, review, verification, deployment, governance) is. The author agrees but argues that spec-driven tools like Anthropic's playbook, Amazon's Kiro, and GitHub's Spec Kit each prescribe a single fixed process, when real organizations need many processes that vary by the risk of a change. A documentation fix, a dependency upgrade, and a payments schema migration should follow different paths with different approvers and evidence requirements. The proposed alternative models each process as a state machine of facts (reviewed, validated, approved) with rules specifying required facts, an automatic-or-human gate, and the permission granted on firing. Classification routes each change to the right risk-based path automatically, process definitions are versioned data reviewed like code, and every transition is logged with who approved it and on what evidence, forming the audit trail as agent autonomy is granted incrementally and expanded as reliability is proven. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why is a single fixed process from tools like Spec Kit or Kiro not enough for AI-generated code changes? A single fixed process forces every change, whether a documentation fix or a payments schema migration, through the same sequence of stages and approvals, regardless of risk. Teams end up routing around the tool for changes that do not fit, making the real process invisible. The alternative is defining process as risk-classified state machines with rules that react to facts like test results and approvals rather than a fixed workflow. _daily.dev surfaces practical takes like this for teams rethinking review and governance around AI-generated code._ ### How can an organization decide which changes need human approval versus automatic merging when using AI coding agents? Autonomy should be granted per action and expanded over time: transitions in a process are set to fire automatically, require approval, or hold, based on demonstrated agent reliability for that class of change. A documentation fix can merge with no human involved once the build passes, while a payments schema migration requires review by a payments owner, validation against production-shaped data, and release approval. _Engineers designing agent approval gates can track evolving practice on daily.dev before shipping their own workflow._ ### What evidence should be required before an AI agent's code change can be approved for production? Evidence must come from systems the agent cannot write to, such as test results and validation in a realistic environment, never from the agent's own self-report. Transitions in the process fire only on these external facts, and each transition is logged with who approved it and on what evidence, forming the audit trail regulators and internal reviewers expect. _daily.dev helps teams building agent review pipelines stay current on evidence-based gating patterns._ ## Similar posts on daily.dev - [Fixing the Verification Gap in the AI-Native SDLC](https://daily.dev/posts/fixing-the-verification-gap-in-the-ai-native-sdlc-7iypalafr) · Mabl Engineering Blog · 10 upvotes · 2 comments - [The AI SDLC transformation playbook](https://daily.dev/posts/the-ai-sdlc-transformation-playbook-hykxezayi) · Atlassian · 1 upvotes · 0 comments - [When code is abundant](https://daily.dev/posts/when-code-is-abundant-gezbqtizx) · GitLab · 0 upvotes · 0 comments - [The AI-Driven SDLC: From Prompting to Process](https://daily.dev/posts/the-ai-driven-sdlc-from-prompting-to-process-bkszru6ln) · Medium · 0 upvotes · 0 comments - [The AI-Native SDLC Starts With Your Infrastructure](https://daily.dev/posts/the-ai-native-sdlc-starts-with-your-infrastructure-2fviwpotm) · MetalBear · 1 upvotes · 1 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#cicd](https://daily.dev/tags/cicd), [#platform-engineering](https://daily.dev/tags/platform-engineering), [#spec-driven-development](https://daily.dev/tags/spec-driven-development) [View this post on daily.dev](https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x) ```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":"The AI-native SDLC won’t be one process","url":"https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x"},"datePublished":"2026-09-12T14:06:31.044Z","dateModified":"2026-09-12T14:13:37.070Z","description":"Anthropic's AI-Native SDLC Playbook argues that with AI agents, code generation is no longer the bottleneck and process (planning, review, verification,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c00eff953c22e91fc81bf9bc76097048?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c00eff953c22e91fc81bf9bc76097048?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The New Stack","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":"The New Stack","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/newstack","url":"https://daily.dev/sources/newstack"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,cicd,platform-engineering,spec-driven-development","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The New Stack","item":"https://daily.dev/sources/newstack"},{"@type":"ListItem","position":3,"name":"The AI-native SDLC won’t be one process"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/the-ai-native-sdlc-won-t-be-one-process-ecjtxeg8x#faq","mainEntity":[{"@type":"Question","name":"Why is a single fixed process from tools like Spec Kit or Kiro not enough for AI-generated code changes?","acceptedAnswer":{"@type":"Answer","text":"A single fixed process forces every change, whether a documentation fix or a payments schema migration, through the same sequence of stages and approvals, regardless of risk. Teams end up routing around the tool for changes that do not fit, making the real process invisible. The alternative is defining process as risk-classified state machines with rules that react to facts like test results and approvals rather than a fixed workflow. daily.dev surfaces practical takes like this for teams rethinking review and governance around AI-generated code."}},{"@type":"Question","name":"How can an organization decide which changes need human approval versus automatic merging when using AI coding agents?","acceptedAnswer":{"@type":"Answer","text":"Autonomy should be granted per action and expanded over time: transitions in a process are set to fire automatically, require approval, or hold, based on demonstrated agent reliability for that class of change. A documentation fix can merge with no human involved once the build passes, while a payments schema migration requires review by a payments owner, validation against production-shaped data, and release approval. Engineers designing agent approval gates can track evolving practice on daily.dev before shipping their own workflow."}},{"@type":"Question","name":"What evidence should be required before an AI agent's code change can be approved for production?","acceptedAnswer":{"@type":"Answer","text":"Evidence must come from systems the agent cannot write to, such as test results and validation in a realistic environment, never from the agent's own self-report. Transitions in the process fire only on these external facts, and each transition is logged with who approved it and on what evidence, forming the audit trail regulators and internal reviewers expect. daily.dev helps teams building agent review pipelines stay current on evidence-based gating patterns."}}]} ```