--- title: This is How Search Systems Work at DoorDash, Instacart,... description: A comparison of how DoorDash, Instacart, and Uber Eats have built their search systems with varying degrees of LLM integration. DoorDash uses LLMs offline to... canonical: https://daily.dev/posts/this-is-how-search-systems-work-at-doordash-instacart-and-uber-eats--tkiitoxak twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: This is How Search Systems Work at DoorDash, Instacart, and Uber Eats! | daily.dev og:description: A comparison of how DoorDash, Instacart, and Uber Eats have built their search systems with varying degrees of LLM integration. DoorDash uses LLMs offline to... og:url: https://daily.dev/posts/this-is-how-search-systems-work-at-doordash-instacart-and-uber-eats--tkiitoxak og:image: https://api.daily.dev/og/posts/tKiItOxaK.png og:image:alt: This is How Search Systems Work at DoorDash, Instacart, and Uber Eats! 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. # This is How Search Systems Work at DoorDash, Instacart, and Uber Eats! **[Pavan Belagatti](https://daily.dev/sources/dfa0u7-fs)** · [@johhnypav](https://daily.dev/johhnypav) · 1 min read · 6 upvotes · 0 comments ## Summary A comparison of how DoorDash, Instacart, and Uber Eats have built their search systems with varying degrees of LLM integration. DoorDash uses LLMs offline to enrich a knowledge graph with an inverted RAG guardrail at runtime. Instacart splits traffic between an offline RAG cache for common queries and a fine-tuned Llama-3-8B model for tail queries. Uber Eats uses a fine-tuned Qwen LLM as an embedding backbone in a two-tower vector search setup. The key takeaway is that LLM integration depth should be driven by existing infrastructure, not just model quality. ## Content What actually happens under the hood when you type "healthy pizza" into DoorDash vs Instacart vs Uber Eats? Let's understand these 3 very different search systems. While they all integrated Large Language Models (LLMs) to solve complex search intents, their architectures look completely different: - **DoorDash**: Keeps retrieval mostly classical. They use LLMs offline to enrich an existing knowledge graph, employing an inverted RAG approach as a guardrail to safely parse queries at runtime. - **Instacart**: Splits the traffic, common queries hit a deeply context-engineered offline RAG cache, while cold-start "tail" queries are handled in real-time by a fine-tuned Llama-3-8B model. - **Uber Eats**: Goes deep with vector search. They fine-tuned a Qwen LLM as the embedding backbone in a two-tower setup, embedding queries in real time to match against pre-computed document vectors. The depth of your LLM integration shouldn't just be about picking the best model, it depends on the infrastructure you already have in place. Like to know more about these search methodologies in depth? This is such an amazing article by ByteByteGo - https://blog.bytebytego.com/p/why-doordash-instacart-and-uber-eats Well, I keep adding so many hands-on engineering videos on my Youtube, subscribe & support: https://www.youtube.com/@pavanbelagatti?sub_confirmation=1 ## Similar posts on daily.dev - [Why DoorDash, Instacart, and Uber Eats Integrated LLMs Into Search Three Different Ways](https://daily.dev/posts/why-doordash-instacart-and-uber-eats-integrated-llms-into-search-three-different-ways-kabch9gjw) · ByteByteGo · 19 upvotes · 2 comments - [Building The Intent Engine: How Instacart is Revamping Query Understanding with LLMs](https://daily.dev/posts/building-the-intent-engine-how-instacart-is-revamping-query-understanding-with-llms-7wcfinvhu) · Instacart · 5 upvotes · 0 comments - [Our Early Journey to Transform Instacart’s Discovery Recommendations with LLMs](https://daily.dev/posts/our-early-journey-to-transform-instacart-s-discovery-recommendations-with-llms-wpv1ryxzx) · Instacart · 1 upvotes · 0 comments - [Using LLMs to build content embeddings for DoorDash search and recommendations](https://daily.dev/posts/using-llms-to-build-content-embeddings-for-doordash-search-and-recommendations-fuy4zkqul) · Doordash · 0 upvotes · 0 comments --- Tags: [#career](https://daily.dev/tags/career), [#llm](https://daily.dev/tags/llm), [#rag](https://daily.dev/tags/rag), [#vector-search](https://daily.dev/tags/vector-search) [View this post on daily.dev](https://daily.dev/posts/this-is-how-search-systems-work-at-doordash-instacart-and-uber-eats--tkiitoxak) ```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":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/this-is-how-search-systems-work-at-doordash-instacart-and-uber-eats--tkiitoxak","headline":"This is How Search Systems Work at DoorDash, Instacart, and Uber Eats!","text":"A comparison of how DoorDash, Instacart, and Uber Eats have built their search systems with varying degrees of LLM integration. DoorDash uses LLMs offline to enrich a knowledge graph with an inverted RAG guardrail at runtime. Instacart splits traffic between an offline RAG cache for common queries and a fine-tuned Llama-3-8B model for tail queries. Uber Eats uses a fine-tuned Qwen LLM as an embedding backbone in a two-tower vector search setup. The key takeaway is that LLM integration depth should be driven by existing infrastructure, not just model quality.","url":"https://daily.dev/posts/this-is-how-search-systems-work-at-doordash-instacart-and-uber-eats--tkiitoxak","datePublished":"2026-08-01T06:15:10.807Z","dateModified":"2026-09-13T21:27:24.662Z","author":{"@type":"Person","name":"Pavan Belagatti","url":"https://daily.dev/johhnypav","image":"https://media.daily.dev/image/upload/s--r9dshbly--/f_auto/v1776930121/avatars/avatar_DFA0U7-fS?_a=BAMAMiWQ0","description":"Developer Evangelist","interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"EndorseAction"},"userInteractionCount":3270}},"image":"https://media.daily.dev/image/upload/s--ydr-pBoV--/f_auto/v1785564911/posts/tKiItOxaK?_a=BAMAMicg0","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":6},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"isPartOf":{"@type":"WebPage","url":"https://daily.dev/sources/dfa0u7-fs","name":"Pavan Belagatti"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Pavan Belagatti","item":"https://daily.dev/sources/dfa0u7-fs"},{"@type":"ListItem","position":3,"name":"This is How Search Systems Work at DoorDash, Instacart, and Uber Eats!"}]} ```