--- title: DoorDash’s Flux Runs 130,000 Engineering Tasks Through... description: DoorDash built an internal cloud platform called Flux to run AI coding agent workloads instead of executing them on developer laptops, addressing limits on... canonical: https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: DoorDash’s Flux Runs 130,000 Engineering Tasks Through Cloud-Based Agents | daily.dev og:description: DoorDash built an internal cloud platform called Flux to run AI coding agent workloads instead of executing them on developer laptops, addressing limits on... og:url: https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt og:image: https://api.daily.dev/og/posts/a53dbpnrt.png og:image:alt: DoorDash’s Flux Runs 130,000 Engineering Tasks Through Cloud-Based Agents 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. # DoorDash’s Flux Runs 130,000 Engineering Tasks Through Cloud-Based Agents **[InfoQ](https://daily.dev/sources/infoq)** · 3 min read · 0 upvotes · 0 comments ## Summary DoorDash built an internal cloud platform called Flux to run AI coding agent workloads instead of executing them on developer laptops, addressing limits on power, uptime, and security. In one month, Flux automated 130,000 engineering tasks and now supports more than 25,000 automated code reviews weekly, with over 300 reusable playbooks and 10,000+ weekly invocations. The architecture rests on four primitives: Firecracker microVM sandboxes for isolation, an in-house MCP gateway (Agent Gateway) for scoped permissions and auditing, YAML-defined playbooks combining agent steps with deterministic code, and multiple invocation surfaces (Slack, GitHub, cron, CLI, conversational interfaces). Sandbox provisioning hits a 95th-percentile SLO of under five seconds. DoorDash expanded Flux beyond code review into CI triage, on-call tasks, maintenance, and ticket-driven development, and moved its Slack integration to public threads for visibility. The move mirrors a broader industry shift toward cloud-based agent execution, exemplified by GitHub Copilot's cloud sandbox support. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is DoorDash's Flux platform and why did they move coding agents off developer laptops? Flux is DoorDash's internal cloud platform for running AI coding agent workloads instead of on developer laptops. It was built because local execution limited available CPU and memory, depended on the developer's device staying connected, and risked giving autonomous agents access to credentials and internal systems already available to the developer, making it harder to monitor agent activity. _Teams weighing local versus cloud execution for coding agents can track approaches like this on daily.dev._ ### How does DoorDash isolate AI agent workloads for security and speed? DoorDash uses Firecracker microVMs to sandbox each agent task, provisioning repositories, dev tools, secrets, and runtime dependencies per environment. Access to internal systems goes through Agent Gateway, an in-house MCP gateway providing scoped permissions and audit logging. Sandbox setup, including cloning repos and installing build tools, hits a 95th-percentile SLO under five seconds. _Engineers designing secure agent sandboxes can follow architecture patterns like this via daily.dev._ ### How many engineering tasks does DoorDash's Flux platform automate? Flux automated 130,000 engineering tasks in a single month, supporting more than 25,000 automated code reviews weekly across over 300 reusable playbooks with more than 10,000 weekly invocations. It expanded from code review into CI triage, on-call tasks, maintenance workflows, and ticket-driven development. _Anyone scaling agent-driven workflows in production can follow adoption benchmarks like these on daily.dev._ ## Similar posts on daily.dev - [Beyond Single Agents: How DoorDash is building a collaborative AI ecosystem](https://daily.dev/posts/beyond-single-agents-how-doordash-is-building-a-collaborative-ai-ecosystem-zfp5vamqj) · Doordash · 1 upvotes · 0 comments --- Tags: [#devops](https://daily.dev/tags/devops), [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp) [View this post on daily.dev](https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt) ```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":"DoorDash’s Flux Runs 130,000 Engineering Tasks Through Cloud-Based Agents","url":"https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt"},"datePublished":"2026-08-31T14:49:27.915Z","dateModified":"2026-08-31T16:27:42.662Z","description":"DoorDash built an internal cloud platform called Flux to run AI coding agent workloads instead of executing them on developer laptops, addressing limits on...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/458ff3623b207e2c8e95b7111bf527f8?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/458ff3623b207e2c8e95b7111bf527f8?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"InfoQ","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":"InfoQ","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/afc3bced3e1e4b188dd9127017a60e0c","url":"https://daily.dev/sources/infoq"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"devops,ai-agents,mcp","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"InfoQ","item":"https://daily.dev/sources/infoq"},{"@type":"ListItem","position":3,"name":"DoorDash’s Flux Runs 130,000 Engineering Tasks Through Cloud-Based Agents"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/doordash-s-flux-runs-130-000-engineering-tasks-through-cloud-based-agents-a53dbpnrt#faq","mainEntity":[{"@type":"Question","name":"What is DoorDash's Flux platform and why did they move coding agents off developer laptops?","acceptedAnswer":{"@type":"Answer","text":"Flux is DoorDash's internal cloud platform for running AI coding agent workloads instead of on developer laptops. It was built because local execution limited available CPU and memory, depended on the developer's device staying connected, and risked giving autonomous agents access to credentials and internal systems already available to the developer, making it harder to monitor agent activity. Teams weighing local versus cloud execution for coding agents can track approaches like this on daily.dev."}},{"@type":"Question","name":"How does DoorDash isolate AI agent workloads for security and speed?","acceptedAnswer":{"@type":"Answer","text":"DoorDash uses Firecracker microVMs to sandbox each agent task, provisioning repositories, dev tools, secrets, and runtime dependencies per environment. Access to internal systems goes through Agent Gateway, an in-house MCP gateway providing scoped permissions and audit logging. Sandbox setup, including cloning repos and installing build tools, hits a 95th-percentile SLO under five seconds. Engineers designing secure agent sandboxes can follow architecture patterns like this via daily.dev."}},{"@type":"Question","name":"How many engineering tasks does DoorDash's Flux platform automate?","acceptedAnswer":{"@type":"Answer","text":"Flux automated 130,000 engineering tasks in a single month, supporting more than 25,000 automated code reviews weekly across over 300 reusable playbooks with more than 10,000 weekly invocations. It expanded from code review into CI triage, on-call tasks, maintenance workflows, and ticket-driven development. Anyone scaling agent-driven workflows in production can follow adoption benchmarks like these on daily.dev."}}]} ```