61
B-Trees: More Than I Thought I'd Want to Know (2021)
B-Trees are not boring, after all
62
Hard problems that reduce to document ranking
There are two claims I’d like to make: LLMs can be used effectively1 for listwise document ranking. Some complex problems can (surprisingly) be solved by transforming them into document ranking problems. I’ve primarily explored both of these claims in the context of using patch diffing to locate N-day…
63
I analyzed chord progressions in 680k songs
And the results surprised me
64
PlanetScale for Postgres is now GA
PlanetScale for Postgres is now generally available.
65
66
TigerBeetle is a most interesting database
This read goes behind the scenes of how TigerBeetle came to be, the incredibly novel software they’ve built, and all of the wacky, wonderful things that make them so special.
67
68
A simple search engine from scratch
*if you include word2vec.
69
The RAG Obituary: Killed by agents, buried by context windows
I’ve been working in AI and search for a decade. First, I built Doctrine, the largest European legal search engine, and now I am building Fintool , an AI-powered financial research platform that helps institutional investors analyze companies, screen stocks, and make investment decisions.
70
71
Exploring PostgreSQL 18's new UUIDv7 support
Exploring what's interesting about UUIDv7 support using a demo crab store.
72
Will Amazon S3 Vectors kill vector databases or save them?
AWS S3 Vectors aims for 90% cost savings for vector storage. But will it kill vectordbs like Milvus? A deep dive into costs, limits, and the future of tiered storage.
73
PSA: SQLite WAL checksums fail silently and may lose data
SQLite WAL has checksums, but on corruption it drops all the data and does not raise error
74
Removing newlines in FASTA file increases ZSTD compression ratio by 10x
First released with Zstandard 1.3.2 in 2017, the --long range match finder increases the compressor’s search window to at least 128MiB, improving deduplication inside large files. This optional feature had substantial performance overheads at launch, but various optimisations have since brought its performance within…
75
Building a Simple Search Engine That Works
You don't need Elasticsearch for most projects. I built a simple search engine from scratch that tokenizes everything, stores it in your existing database, and scores results by relevance. Dead simple to understand and maintain.
76
Reverse geocoding is hard
My wife and I run OpenBenches - a crowd-sourced database of nearly 40,000 memorial benches. Every bench is geo-tagged with a latitude and longitude. But how do you go from a string of digits to something human readable? How do I turn -33.755780,150.603769 into "42 Wallaby Way, Sydney, Australia"? Luckily, that's a…
77
Data on AI-related Show HN posts
The idea for this article didn't exist in my brain before this morning. But there I was, scrolling the New page and even more tired of all the AI-related Show HN posts than usual. I was confident that their numbers were multiplying and wanted proof. Exactly how much more AI crap is on my lawn compared to last year?
78
How we replaced Elasticsearch and MongoDB with Rust and RocksDB
Radar offers geocoding, place search, and fraud detection APIs, powered by HorizonDB, a Rust-based engine delivering 20k QPS and sub-millisecond latency.
79
The power of interning: making a time series database smaller
This week-end project started by browsing the open-data repository of Paris’ public transport network, which contains various APIs to query real-time departures, current disruptions, etc.The data reuse section caught my eye, as it features external projects that use this open data.In particular, the RATP status…
80
Composable SQL
Better SQL through typed, composable query fragments.
81
Are hard drives getting better?
After 13 years, the data is clear—the bathtub curve does not apply to hard drives. And, more importantly, hard drives are failing less and lasting longer.
82
83
Understanding Smallpond and 3FS
A concise breakdown of what smallpond and 3FS are, how to use them, and if they're suitable for your data processing needs.
84
We cut our Mongo DB costs by 90% by moving to Hetzner
We moved MongoDB Atlas off AWS and onto self-hosted Hetzner boxes — 90% cost cut, no performance loss. Here's the setup, the Ansible, and what broke.
85
86
Polars Cloud: The Distributed Cloud Architecture to Run Polars Anywhere
We introduce Polars Cloud, the distributed cloud architecture to scale and run Polars anywhere.
87
MongoBleed Explained Simply
CVE-2025-14847 allows attackers to read any arbitrary data from the database's heap memory. It affects all MongoDB versions since 2017, here's a simple explanation:
88
Decrypting encrypted files from Akira ransomware using a bunch of GPUs
I recently helped a company recover their data from the Akira ransomware without paying the ransom. I'm sharing how I did it, along with the full source code. Update: since this article was written, a new version of Akira ransomware has appeared that can't be decrypted with this method The c
89
Satellogic's Open Satellite Feed
Benchmarks & Tips for Big Data, Hadoop, AWS, Google Cloud, PostgreSQL, Spark, Python & More...
90
Shardines: SQLite3 Database-per-Tenant with ActiveRecord
There is a pattern I am very fond of - “one database per tenant” in web applications with multiple, isolated users. Recently, I needed to fix an application I had for a long time where this database-per-tenant multitenancy utterly broke down, because I was doing connection management wrong. Which begat the question…