Hacker News: pjot

New comment by pjot in "I've operated petabyte-scale ClickHouse clusters for 5 years"

pjot — Sat, 12 Sep 2026 00:20:10 +0000

I don’t think it’s necessarily the use case changed but maybe the constraints. Streaming directly to Clickhouse is a great solution. Recent approaches I’ve seen though buffer data as parquet in object storage first, and read data from there. Im not sure if CH would be the preference

New comment by pjot in "Project Blinkenlights"

pjot — Fri, 11 Sep 2026 23:01:29 +0000

Thought this was the ascii Star Wars in your terminal à la

  telnet towel.blinkenlights.nl

New comment by pjot in "Discovery of a new OpenAI agent message board"

pjot — Fri, 04 Sep 2026 19:48:10 +0000

1008 pages, wow!

New comment by pjot in "Defrag98: Windows 98 Disk Defragmenter Simulator Online"

pjot — Sun, 30 Aug 2026 02:04:33 +0000

It also made a really satisfying “boyoyoyinnng” sound

New comment by pjot in "Defrag98: Windows 98 Disk Defragmenter Simulator Online"

pjot — Sat, 29 Aug 2026 23:36:13 +0000

Will need to degauss after this

New comment by pjot in "Nitter and XCancel receive cease and desist notices"

pjot — Tue, 25 Aug 2026 21:22:14 +0000

Where do you get this free beer?

New comment by pjot in "Gmail might partially be to blame for receiving emails from other Sean Conners"

pjot — Mon, 17 Aug 2026 04:31:09 +0000

I used to love watching the “unlimited” storage counter increase as each second went by

New comment by pjot in "Asynchronous I/O in DuckDB: Work, Thread, Work"

pjot — Sun, 16 Aug 2026 04:49:05 +0000

I wonder how this would work in trying to parallelize the worker threads (multiple duckdb instances) coordinating them via Quack.

Ducks all the way down!

New comment by pjot in "Why Large Language Models Fail at Tabular Prediction"

pjot — Tue, 04 Aug 2026 15:04:14 +0000

One step further are those who want to point an llm directly at the data warehouse to get the data needed to run predictions

New comment by pjot in "Systems and Delays"

pjot — Sun, 26 Jul 2026 03:47:43 +0000

New comment by pjot in "Alternate Clock Designs and Time Systems"

pjot — Fri, 10 Jul 2026 17:22:55 +0000

Here’s a different kind of binary clock https://www.hey.earth/posts/binary-clock

New comment by pjot in "Anthropic Urges Global Pause in AI Development, Flags 'Self-Improvement' Risk"

pjot — Fri, 05 Jun 2026 13:50:45 +0000

The actual post from Anthropic: https://www.anthropic.com/institute/recursive-self-improveme...

New comment by pjot in "1-Click GitHub Token Stealing via a VSCode Bug"

pjot — Wed, 03 Jun 2026 05:34:41 +0000

  > created bunch of fake spam pages and trying to mine crypto

Pages like GitHub pages? We’re repos being created in your account? Curious how you discovered that your tokens were pwned

New comment by pjot in "Love systemd timers"

pjot — Tue, 02 Jun 2026 16:23:20 +0000

After years of using orchestration tools like airflow and dagster so many lightbulbs have just lit up in my head.

I wish documentation for tools would explain their abstractions concepts in terms of its primitives.

Great post, thanks!

New comment by pjot in "Ebola Outbreak Now Third Largest Recorded and "Spreading Rapidly""

pjot — Sat, 23 May 2026 17:47:26 +0000

  > The Chinese are a decade ahead of the west when it comes to building cars.

Is this true? From years of watching Top Gear any Chinese car that was tested was laughably bad.

New comment by pjot in "NanoTDB – Golang Append-Only Time Series DB"

pjot — Sat, 16 May 2026 00:49:13 +0000

DuckDB has indexes. One specifically for highly selective point lookups. The structure is called an adaptive radix tree. The workloads you’re describing are sort of duckdb’s bread and butter (even the parquet stuff it’ll handle out of the box)

https://duckdb.org/2022/07/27/art-storage

New comment by pjot in "Pretty Fish: A better mermaid diagram editor"

pjot — Wed, 15 Apr 2026 15:20:08 +0000

To be fair, pie charts are horrible in general.

New comment by pjot in "Show HN: Hacker News archive (47M+ items, 11.6GB) as Parquet, updated every 5m"

pjot — Thu, 19 Mar 2026 02:56:34 +0000

Re-reading your comment :) Yes, my demo has just a simple loop when loading the embeddings.

I was replying more towards the latency you mentioned. Because duckdb runs on device, you save yourself the additional round trip network time when comparing similarities.

New comment by pjot in "Show HN: Hacker News archive (47M+ items, 11.6GB) as Parquet, updated every 5m"

pjot — Thu, 19 Mar 2026 01:51:24 +0000

I did this but used duckdb as the vector store. Works really well, quite fast too.

https://github.com/patricktrainer/duckdb-embedding-search

New comment by pjot in "Using an engineering notebook"

pjot — Thu, 12 Feb 2026 04:47:53 +0000

For me, it helps to slow down my thoughts and aides deep work. I draw diagrams, connect blurbs with arrows, and “link” to other page numbers.