Stu is the mastermind behind much of ParadeDB's internals. Highly highly recommend a watch.

PGConf.dev 🐘@pgconf.dev · 3w ago

We've posted a video of Stu Hood's talk Pushing the Limits of the Index API: Building a Columnar Store without a TAM https://youtu.be/VPZ... #postgresql #PGConfdev

> Why search and analytics are now converging From Elastic's latest product announcement^ I've been saying this since the beginning of ParadeDB. An incredibly exciting time where data infra sprawl is collapsing. The future will be simpler, cheaper, and more powerful software for all.

ParadeDB Cloud coming very soon!!

ParadeDB@paradedb.com · 2w ago

You have a product, and it needs search. You have agents, and they need retrieval. We're building ParadeDB Cloud for you, so you can focus on shipping: one Postgres for your data, full-text search, vector retrieval, and aggregations. Waitlist open now: paradedb.com/cloud/

This graph really says it all

The Apache DataFusion 54.0.0 release (2 months of development had 139 distinct contributors. 🤯 github.com/apache/dataf...

ParadeDB is now natively available in Postgres.app for all macOS users out there! postgresapp.com/extensions/

We wrote a benchmarking tool to compare ourselves to state-of-the-art search/OLAP databases: github.com/paradedb/ben... If you're familiar with any of them, we'd love your help auditing our config files to ensure they are as optimized as possible (to put up a good fight!)

GitHub - paradedb/benchmarker: Database benchmarking suite based on Grafana K6. Benchmark ParadeDB, Elasticsearch, OpenSearch, ClickHouse, and others!

Database benchmarking suite based on Grafana K6. Benchmark ParadeDB, Elasticsearch, OpenSearch, ClickHouse, and others! - paradedb/benchmarker

github.com

ParadeDB is hiring a distributed systems engineer to come build our ParadeDB Cloud with us. This is a high-impact role that could lead our entire platform initiative. More details here: paradedb.notion.site/Distributed-...

More integrations coming to ParadeDB! We now support @DrizzleORM natively for ParadeDB indexes! This adds to Django, SQLAlchemy, Rails (ActiveRecord) and EF Core (C#) for the now-supported list of clients for ParadeDB.

Benchmarks lie to you. But if you build your benchmarks for yourself, the incentives change. At ParadeDB, we're first trying to convince ourselves. If we do, we think we'll also convince others. This post is the story of how the same question leads to 3 results, and how we optimize for each.

ParadeDB@paradedb.com · 2mo ago

To explore how methodology can color results we used our open-source benchmarker to compare ParadeDB and Postgres FTS three times. - P1 (1 term, closed loop): within 10% - P2 (40-term rotation): 29x throughput - P3 (open loop @ 50 QPS): 47x P99 www.paradedb.com/blog/benchma...

One of the most common question we get for ParadeDB is "When will it be hosted?". With this partnership, the answer is now. You can now host ParadeDB on Render with the click of a button, so you can focus on building your app.

ParadeDB@paradedb.com · 2mo ago

ParadeDB and Render have teamed up to offer one-click deployments. www.paradedb.com/blog/render

This is *so cool*: med.stanford.edu/news/all-new... Software is cool, but man, biology is something else. I really hope this discovery works as well as they say here. It would be life-changing

Inhibiting a master regulator of aging regenerates joint cartilage in mice

Osteoarthritis due to aging or injury affects 20% of the U.S. population. A Stanford Medicine-led study found that blocking a ‘gerozyme’ reverses cartilage loss in mice and human tissue.

med.stanford.edu

Datadog is investing in open source and Tantivy! This is an incredible opportunity for anyone interested in data infra. You'll get to work on open source at Datadog and work with the incredible Paul Masurel, François Massot, and Adrien Guillo. careers.datadoghq.com/detail/75555...

ParadeDB is hiring a platform/distributed systems engineer to come build ParadeDB Cloud with us. This is a high leverage role -- will get to make a lot of big technical decisions and report directly to me. North America only, strong preference for west coast. DMs open.

1/4. ParadeDB is fast. But (until now), you've had to either try it yourself or take our word for it. We're long overdue for some public-facing benchmarks! Most companies out there selectively pick the workloads that make them look good. We didn't want to do that. This is why we built this project.

ParadeDB@paradedb.com · 3mo ago

Today we open sourced @paradedb.com Benchmarker, a workload agnostic benchmarking runner built on top of #grafana #k6. Supports PostgreSQL, Elasticsearch, OpenSearch, ClickHouse, MongoDB, and ParadeDB. blog: paradedb.com/blog/what-we... repo: github.com/paradedb/ben...

ParadeDB is working on vector search. Our implementation is designed to solve pgvector's limitations and integrate nicely with our existing full-text search features. We have room for one more engineer. If you've shipped production vector search before, I'd love to talk. DMs open.

One of the most common question we get for ParadeDB is "When will it be hosted?". With this partnership, the answer is now. You can now host ParadeDB on @railway.com with the click of a button, so you can focus on building your app. railway.com/deploy/parad...

Deploy ParadeDB

Deploy ParadeDB to the cloud for free with Railway, the all-in-one intelligent cloud provider. Simple, Elastic-quality search for Postgres

railway.com

ParadeDB@paradedb.com · 3mo ago

ParadeDB and Railway have teamed up to offer one-click deployments. www.paradedb.com/blog/railway

The Cofactr team got burned by MongoDB. They needed a database that was 1) reliable, 2) operationally simple and 3) could be self-hosted within AWS GovCloud. ParadeDB was a perfect fit, enabling them to consolidate search + OLTP in a single datastore.

ParadeDB@paradedb.com · 3mo ago

We’re proud to partner with Cofactr to power intelligent sourcing and search across their end-to-end electronics purchasing platform. www.paradedb.com/customers/ca...

The incredible Isaac Van Doren from our team is organizing Software Should Work, an independent conference about software reliability, on July 16-17. ParadeDB is not affiliated, although we are proud sponsors. Speakers include the creators of Zig, SQLite, etc. More info: softwareshould.work

Software Should Work

An independent conference on software reliability in Columbia, MO on July 16-17, 2026

softwareshould.work

ParadeDB hosted on Render 👀

Next stop: 10 millions

ParadeDB@paradedb.com · 3mo ago

The ParadeDB project has officially reached one million Docker pulls. We're grateful to all our users and customers for the journey so far.

A very brief and entertaining read: dpc.pw/posts/i-dont... We still love community contributions, but it is true that the bottleneck is understanding, not writing code.

The slides from my talk are now live at postgresconf.org/conferences/...

pg_search: Bringing Elasticsearch-Grade Search to PostgreSQL

This is a talk about search. We will cover search use cases, Postgres' existing search functionality, and areas where it needs improvement. We'll then discuss pg_search, a Postgres extension that brin...

postgresconf.org

Philippe Noël@philippemnoel.bsky.social · 4mo ago

I'll be speaking at PostgresConf San Jose tomorrow (April 21). Short notice, but here you go: postgresconf.org/conferences/...

I'll be speaking at PostgresConf San Jose tomorrow (April 21). Short notice, but here you go: postgresconf.org/conferences/...

pg_search: Bringing Elasticsearch-Grade Search to PostgreSQL

This is a talk about search. We will cover search use cases, Postgres' existing search functionality, and areas where it needs improvement. We'll then discuss pg_search, a Postgres extension that brin...

postgresconf.org

I'm thrilled about this case study coming out. The Modern Treasury team have been an incredible design partner from the early days. They had very strict requirements around data freshness, and we're honoured to be the database powering many of their user-facing dashboards and endpoints.

ParadeDB@paradedb.com · 4mo ago

We’re proud to be partnering with Modern Treasury to deliver fast, accurate search across their suite of money movement UIs and APIs. www.paradedb.com/customers/ca...

1/3. A frequent question I get about ParadeDB is "Can it support distributed search?" Elasticsearch is sharded, after all? Starting today, the answer is yes: distributed search is now possible in Postgres thanks to ParadeDB x Citus.

What shocked me the most in the AI wave so far is seeing many smart people start to advocate for quantity over quality. In a world where so many can "make", quality is so clearly the defining factor. In fact, it seems obvious that AI will raise the bar for quality, not lower it.

The ParadeDB Community Slack has just reached 1,000 members! The most unexpected part is certainly the growth in external contributors, which has been really wonderful to witness. We are deeply thankful to everyone who contributed.

Stu's talk is now available for anyone who couldn't attend South Bay Systems to watch: youtu.be/TeFsBVIYBis?... I highly recommend watching it. It's ~30 minutes and goes into details on how ParadeDB achieves faster-than-Postgres JOINs.

Generalized Consensus & ​Native Top-K Joins in ParadeDB

YouTube video by South Bay Systems

youtu.be

ParadeDB@paradedb.com · 5mo ago

On March 31st, @stuhood.sh will be speaking about how ParadeDB performs native Top K JOINs at South Bay Systems. We look forward to seeing you there: luma.com/2g3exvjw?tk=...

1/4. Today, a bit of a different blog post. James from our team sat down with @fulmicoton.bsky.social, the main author of Tantivy, to discuss his journey and how he created such a world-changing project. 🧵

ParadeDB@paradedb.com · 4mo ago

Full-text search was solved in 2016, Lucene had owned the space for a decade. Rewriting it in Rust, as a side project, on a plane sounds like a terrible idea. Enter Paul Masurel and Tantivy ... www.paradedb.com/blog/tantivy...