Lots of talk on using AI for writing so here is a 100% human written blog post on how I started writing more without using AI for help. tylerhillery.com/blog/writing...
Tyler Hillery
@tylerhillery.com
Software Engineer (Storage) @Supabase Creator of pypacktrends.com Co-organizer of madisonsystems.xyz Peeling back the layers of abstraction Learn more about me: tylerhillery.com
Supabase is now deploying >1m new Postgres databases every week & that doesn't include read replicas
The next Madison Systems paper discussion is scheduled. This time at the Memorial Union Terrace to enjoy the WI summer. Register Here: luma.com/xdmbcw2t?tk=...
High signal to me is seeing a company landing page that doesn't have "AI" in it. I think @turbopuffer.bsky.social is a great example of this. You'd think a vector DB would be all over "AI" but nope. Just their core value backed up by metrics.
Gergely Orosz@gergely.pragmaticengineer.com · 4mo agoWild to see every VC funded and publicly traded tech company rebrand to being an AI company Paging company —> the platform for AI-first operations CDN —> AI Cloud Helpdesk software —> AI-powered helpdesk platform o11y —> AI-powered intelligent o11y Everywhere…
This Tuesday, Hannes announced Quack, our new protocol that turns DuckDB into a client-server database. Watch his talk, recorded at the AI Council, where he explained how Quack works in practice, what stack it's built on, how it performs, and what our long-term ambitions are. youtu.be/L_lttD-d1wc
DuckDB-Quack announcement at AI Council
YouTube video by DuckDB
youtube.com
Reminder, Madison Systems is having another paper discussion in 2 weeks. This time, we'll be discussing: Scalability! But at what COST? luma.com/1enx2p3m
Huh, this is cool. I’d never really thought through why there’s a gap between the upper and lower case ranges in ASCII. tylerhillery.com/blog/why-don...
Why Don’t Lowercase Letters Come Right After Uppercase Letters in ASCII? – Tyler Hillery
tylerhillery.com
I wrote about a particular part of ASCII that finally clicked for me today tylerhillery.com/blog/why-don...
What bogus, I always thought these numbers were tied to physical measurements
Man I should have switched pypacktrends over to slot based pricing a while ago. Seems like once you get over the > 1TB scanned per month, bytes processed based billing isn't worth it
I wrote about why I like computers: tylerhillery.com/blog/link-bl...
I like computers – Tyler Hillery
tylerhillery.com
You know it's serious when you bust out the debugger and start drawing diagrams
Finally pulled the trigger on Math Academy. I haven’t taken a math course since Calc 2 in 2017, which was so hard for me I changed majors. My goal is to follow the foundations path through Discrete Mathematics.
Root causing a bug is one of the greatest feelings as a software engineer
Great discussion today! A surprising amount of Ceph mentions and I had to bring up the napkin math repo a few times
Tyler Hillery@tylerhillery.com · 6mo agoMadison Systems is having another paper discussion this Saturday on The Google File System. Hope to see you there! luma.com/zfh0lcxf
Madison Systems is having another paper discussion this Saturday on The Google File System. Hope to see you there! luma.com/zfh0lcxf
I contributed SHOW CLIENTS metrics to the Prometheus PgBouncer exporter. Similar to pg_stat_activity, but for PgBouncer. Now you can monitor client connections over time. Wrote a bit about it: tylerhillery.com/blog/monitor...
New blog post about the most important thing I've learned while ramping up on Node.js: Don't block the event loop. tylerhillery.com/blog/tyler-t...
I noticed this SQLite query averaging ~90ms event though package_name is the PK, the table has ~865k rows and is ~166MB. Param is a prefix pattern like 'duck%'. Any guesses why it was still that slow?
