---
title: "Building Ask DoorDash (Part 2): Intelligence"
url: https://daily.dev/posts/building-ask-doordash-part-2-intelligence-4xcmvaewz
source_url: https://careersatdoordash.com/blog/building-ask-doordash-part-two-intelligence
type: article
source: "Doordash"
published: 2026-06-18T21:54:49.998Z
updated: 2026-06-19T04:20:25.250Z
tags: ["llm", "ai-agents", "vector-search"]
reading_time: 18
upvotes: 2
comments: 0
language: 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.

# Building Ask DoorDash (Part 2): Intelligence

**[Doordash](https://daily.dev/sources/doordash)** · 18 min read · 2 upvotes · 0 comments

## Summary

DoorDash's engineering team details the agent memory system powering Ask DoorDash, their agentic ordering assistant. The system bridges offline long-term memory (behavioral preferences, brand affinities, dietary patterns) with live agents through a three-layer architecture: offline memory generation, distributed SQL+vector storage, and an orchestration tooling layer. Key mechanisms include task-aware query planning across semantic and keyword search, multi-tenant namespace partitioning, and a conversational memory flywheel that asynchronously extracts durable facts from chat sessions and writes them back to long-term storage. The system applies domain-aware durability rules (e.g., 'I don't want ramen' is transient in restaurant context but brand preferences in grocery are captured immediately), two-stage semantic deletion to avoid over-deleting adjacent memories, and fact-lifecycle management based on what a fact is about rather than uniform TTL decay. Early production results show grocery sessions with memory converted to checkout at ~24% higher rates, basket size increased ~17%, and intent misunderstandings dropped ~33% compared to sessions without memory.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://careersatdoordash.com/blog/building-ask-doordash-part-two-intelligence>

## Similar posts on daily.dev

- [How DoorDash Built an AI Shopping Assistant That Doesn’t Rely on the LLM Alone](https://daily.dev/posts/how-doordash-built-an-ai-shopping-assistant-that-doesn-t-rely-on-the-llm-alone-w3gyrfrce) · InfoQ · 5 upvotes · 1 comments
- [Building a unified consumer memory for personalization at scale](https://daily.dev/posts/building-a-unified-consumer-memory-for-personalization-at-scale-emiywz2l0) · Doordash · 0 upvotes · 0 comments
- [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: [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#vector-search](https://daily.dev/tags/vector-search)

[View this post on daily.dev](https://daily.dev/posts/building-ask-doordash-part-2-intelligence-4xcmvaewz)