Sulat.com / Writing on AITuesday, November 17, 2026
Agent Engineering/17 November 2026

Perplexity’s CEO says its DynamoDB replacement saves tens of millions a year

Aravind Srinivas told the Newcomer podcast the in-house database saves several tens of millions of dollars a year. Perplexity's engineering post said at least 20%; his September X post said up to US$100 million.

A glowing 3D Perplexity logo set into a cobblestone arch while small robots on scaffolding lay more stones, with an empty orange crate in the foreground.

At a glance

  • Perplexity CEO Aravind Srinivas told the Newcomer podcast, in an episode published 13 November, that a database built with Perplexity's own coding agents replaced AWS DynamoDB and saves "several tens of millions of dollars a year".
  • Perplexity's 14 September engineering post put the saving at "at least 20%" against DynamoDB, and Srinivas's X post on 15 September said "up to a hundred million dollars yearly"; neither comes with a published DynamoDB bill.
  • CobbleDB is a roughly 40,000-line Rust key-value store that Perplexity says two engineers and hundreds of coding agents built in two months; median batch-read latency fell from 31.4 ms to 5.60 ms.
  • Perplexity says engineers set the architecture and authorised production operations while agents reviewed changes, prepared fixes and tests, and monitored rollouts, and that it plans to open-source CobbleDB soon.

Perplexity CEO Aravind Srinivas says the in-house database that replaced AWS DynamoDB in Perplexity’s search stack is “saving us like several tens of millions of dollars a year”, on the Newcomer podcast. The episode went up on 13 November.

Srinivas said Perplexity “completely replaced” its use of DynamoDB, “which is where we hosted our entire trillion web page index”, and built the new database “from scratch using our own coding agents”. He raised it as an example of using agents to cut infrastructure costs. The database is CobbleDB, which Perplexity described in an engineering post on 14 September.

Three figures for one project

The podcast figure is the third Perplexity has attached to CobbleDB. The engineering post gave a percentage, Srinivas’s X post gave a ceiling in dollars, and the podcast gave a range. None of them comes with a published DynamoDB bill.

DateWhereFigureWhat it rests on
14 Sep 2026Perplexity engineering postAt least 20% cheaper than DynamoDBAn internal cost model of storage, write and read capacity at every commitment tier
15 Sep 2026Srinivas on X“up to a hundred million dollars yearly”No breakdown in the post
13 Nov 2026Newcomer podcast“several tens of millions of dollars a year”No breakdown in the interview
Sources: Perplexity’s engineering post, Srinivas’s X post and the Newcomer podcast.

The engineering post puts no dollar amount on the saving. It says the real savings are “likely even higher” than 20%, because the estimate leaves out expected backup savings from compression.

Three stat cards: 20%-plus cheaper than DynamoDB, up to US$100M a year, and several tens of millions a year
Perplexity has given three different savings figures for one database.

What the engineering post describes

CobbleDB is a distributed key-value store, a database that hands back a stored record when you give it a key. Its keys are hashed page URLs. Its values are each page’s processed passages plus a vector embedding for each passage, the numerical representation used to match it to a query. The post says the core is roughly 40,000 lines of Rust, built in two months by two engineers and hundreds of coding agents.

Before the move, a prepared-pages table in DynamoDB held the passages and embeddings for query-time reads, with full page content and metadata stored separately. Perplexity says DynamoDB charged for every byte read or written, and that a single Search API request fetches 100 to 120 pages at about 50 KB each. The post doesn’t give a page count for the index.

Median batch-read latency fell from 31.4 ms to 5.60 ms, and the slowest 1% of reads fell from 123 ms to 24.2 ms. Perplexity says both sets of numbers come from live traffic at about 200,000 requests per second, measured at different times, so they aren’t matched requests.

Bar chart: median batch read 31.4 ms on DynamoDB vs 5.60 ms on CobbleDB; slowest 1% 123 ms vs 24.2 ms
Perplexity’s live numbers show reads about five times faster on CobbleDB.

Where the agents fit

The post says the agents reviewed code and infrastructure changes, prepared fixes, tests and monitoring changes, tracked CI and review gates, and watched rollouts and backup and restore exercises. Engineers set the architecture, reviewed consequential changes and authorised production operations, and Perplexity says it kept production actions “explicitly human-owned”.

Two other pieces of software sit around CobbleDB. Pillar keeps versioned page state on cheap disks, and Lorry packs updates into batches. CobbleDB, which runs on faster NVMe storage, ingests those batches and serves the pages Perplexity wants in search results.

Three stacked layers: Pillar keeps page versions, Lorry packs update batches, CobbleDB serves page records
Updates now flow through three parts instead of writing straight into DynamoDB.

Perplexity says it plans to open-source CobbleDB “soon”. It hasn’t given a date.


Discover more from Sulat AI & Tech News

Subscribe to get the latest posts sent to your email.