--- title: How AI is transforming analytics at Grab | daily.dev description: Grab's engineering team shares how they've built a five-level autonomy ladder for AI-driven analytics, moving from AI-assisted work (L2) to fully autonomous... canonical: https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How AI is transforming analytics at Grab | daily.dev og:description: Grab's engineering team shares how they've built a five-level autonomy ladder for AI-driven analytics, moving from AI-assisted work (L2) to fully autonomous... og:url: https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome og:image: https://api.daily.dev/og/posts/c5954FOMe.png og:image:alt: How AI is transforming analytics at Grab 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 AI is transforming analytics at Grab **[Grab Tech Blog](https://daily.dev/sources/grab)** · 13 min read · 4 upvotes · 1 comments ## Summary Grab's engineering team shares how they've built a five-level autonomy ladder for AI-driven analytics, moving from AI-assisted work (L2) to fully autonomous loops (L5). Key internal systems include Spartan, an agentic Slack-based analytics workflow with 50+ skills and 120+ analysis frameworks; Scarlet, a near-self-healing pipeline tool; ContextIQ for managing metric context lifecycles; and BriX, an internal portal that grew 10x since September 2025. Concrete results include mechanical ticket share dropping from 44% to 30%, cycle times reduced by ~33%, and self-serve agent resolution of SQL requests rising from 50% to 81%. The post argues that as agents own more of the analytics loop, analysts shift from producing artifacts to owning problem framing, judgment, and decisions. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the five-level autonomy ladder Grab uses for AI in analytics? It ranges from L2 (AI-assisted, human executes every step) through L3 (human plans, agent owns steps, human reviews) and L4 (agent plans and owns workflows, human reviews only at gates) up to L5 (end-to-end autonomous, human sets objectives and reviews exceptions only). Grab modeled the ladder on Dan Shapiro's framing of five levels for AI coding, adapted for analytics loops. _Teams designing agent autonomy levels for their own workflows can track frameworks like this one on daily.dev._ ### What capabilities are needed to move an analytics workflow up an AI autonomy ladder? Five capabilities gate the climb in order: execution (a stack that runs the loop end to end), knowledge (certified metrics with discoverable, agent-readable context), control (mechanical checks for repeatable expectations), review and governance (agents checking their own output against gates), and learning (encoding fixes into context docs, golden datasets, and evals when an agent fails the same way twice). _Engineers scoping agentic workflow platforms can follow real capability breakdowns like this on daily.dev._ ### What measurable impact did Grab see from shifting analytics work to AI agents? The share of mechanical analyst tickets (data prep, alerting, reporting) fell from 44% in February to 30% by June, and cycle times dropped about 33%. In a self-serve Slack channel, fully automated answers rose from 53% to 67% for metric questions, 63% to 90% for data pulls, and 50% to 81% for SQL requests between March and May. _Anyone weighing agentic analytics investment against real outcomes can find case studies like this on daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@agustinbarrientos** · 0 upvotes > What tells you an analytics workflow can move from L3 to L4, especially when review quality is harder to measure than task completion? ## Similar posts on daily.dev - [Grab Cuts Mechanical Analytics Work From 44% to 30% with AI Agents](https://daily.dev/posts/grab-cuts-mechanical-analytics-work-from-44-to-30-with-ai-agents-m43v7awpy) · InfoQ · 0 upvotes · 0 comments - [How Grab is Using AI Agents to Boost Team Productivity](https://daily.dev/posts/how-grab-is-using-ai-agents-to-boost-team-productivity-fkppsh0tc) · ByteByteGo · 2 upvotes · 0 comments - [From firefighting to building: How AI agents restored our team’s core productivity](https://daily.dev/posts/from-firefighting-to-building-how-ai-agents-restored-our-team-s-core-productivity-zfyvdslnl) · Grab Tech Blog · 2 upvotes · 0 comments - [The Oversight Ladder: the five levels of AI coding autonomy](https://daily.dev/posts/the-oversight-ladder-the-five-levels-of-ai-coding-autonomy-ds0q5sfsf) · Medium · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#data-analysis](https://daily.dev/tags/data-analysis), [#data-engineering](https://daily.dev/tags/data-engineering) [View this post on daily.dev](https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome) ```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 AI is transforming analytics at Grab","url":"https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome"},"datePublished":"2026-08-03T01:34:50.491Z","dateModified":"2026-09-14T08:31:12.503Z","description":"Grab's engineering team shares how they've built a five-level autonomy ladder for AI-driven analytics, moving from AI-assisted work (L2) to fully autonomous...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/959da123bc570c3c8ba2b797d57fed1a?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/959da123bc570c3c8ba2b797d57fed1a?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Grab Tech Blog","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":"Grab Tech Blog","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/c8b8850354ee4494a05777b62e42cc62","url":"https://daily.dev/sources/grab"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":4},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"llm,ai-agents,data-analysis,data-engineering","timeRequired":"PT13M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Grab Tech Blog","item":"https://daily.dev/sources/grab"},{"@type":"ListItem","position":3,"name":"How AI is transforming analytics at Grab"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome","comment":[{"@type":"Comment","text":"What tells you an analytics workflow can move from L3 to L4, especially when review quality is harder to measure than task completion?","datePublished":"2026-08-03T16:50:23.603Z","url":"https://daily.dev/posts/c5954FOMe#c-TY4vMLC3J","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-ai-is-transforming-analytics-at-grab-c5954fome#faq","mainEntity":[{"@type":"Question","name":"What is the five-level autonomy ladder Grab uses for AI in analytics?","acceptedAnswer":{"@type":"Answer","text":"It ranges from L2 (AI-assisted, human executes every step) through L3 (human plans, agent owns steps, human reviews) and L4 (agent plans and owns workflows, human reviews only at gates) up to L5 (end-to-end autonomous, human sets objectives and reviews exceptions only). Grab modeled the ladder on Dan Shapiro's framing of five levels for AI coding, adapted for analytics loops. Teams designing agent autonomy levels for their own workflows can track frameworks like this one on daily.dev."}},{"@type":"Question","name":"What capabilities are needed to move an analytics workflow up an AI autonomy ladder?","acceptedAnswer":{"@type":"Answer","text":"Five capabilities gate the climb in order: execution (a stack that runs the loop end to end), knowledge (certified metrics with discoverable, agent-readable context), control (mechanical checks for repeatable expectations), review and governance (agents checking their own output against gates), and learning (encoding fixes into context docs, golden datasets, and evals when an agent fails the same way twice). Engineers scoping agentic workflow platforms can follow real capability breakdowns like this on daily.dev."}},{"@type":"Question","name":"What measurable impact did Grab see from shifting analytics work to AI agents?","acceptedAnswer":{"@type":"Answer","text":"The share of mechanical analyst tickets (data prep, alerting, reporting) fell from 44% in February to 30% by June, and cycle times dropped about 33%. In a self-serve Slack channel, fully automated answers rose from 53% to 67% for metric questions, 63% to 90% for data pulls, and 50% to 81% for SQL requests between March and May. Anyone weighing agentic analytics investment against real outcomes can find case studies like this on daily.dev."}}]} ```