Ellie Huxtable
@ellie.wtf
5 months ago huge compliment to be compared with jj like this 🫶
💬 3
♻️ 0
❤️ 9
huge compliment to be compared with jj like this 🫶
GA4 filters out bot traffic by design. That includes the bots you now want to see: GPTBot, ClaudeBot, PerplexityBot.
Nahrin Jalal on how Netlify Log Drains + Profound give you the AEO visibility your analytics stack was built to hide.
www.netlify.com/blog/trackin... alright, I'm doing it - trying to make "abottom" happen 1. it's shorter than "at the bottom" 2. people will get it, because "atop" let's get this thing in the dictionary
Everyone said AI design was hopeless. We proved them wrong.
Meet /design avantis/dimensional/maybe others have total-market-like etfs that delay ipo inclusion for longer (for ex: avus). extremely not investing advice, they're somewhat higher expense ratio, i don't use them but i am thinking about the same stuff.
hard to overstate how much i hate ai scrapers, what this article describes is one thousand percent true and i bet lots and lots of engineers at other companies are battling the same enemy right now weirdgloop.org/blog/clankers At scale, you can only put off sharding for so long. YouTube scaled MySQL with Vitess, and now we're building the same for Postgres.
i've spent ~7ish years doing things ink+switch adjacent, previously at observablehq which had some of the same coder-catnip aspirations (reactive graph-based programming) and i can tell you that it's probably a lot nicer to run a self-funded research lab on those concepts than to try and make money
the initial version did have replayable execution, pseudo-in-memory durable storage, etc. we removed all of that because interop mattered a lot more (membrane.io is closer to that vision in terms of current software, or durable objects). Check out elicit.com/blog/system... for recent updates to Elicit Systematic Review. In rare disease, every paper counts towards informing research questions. Patient cohorts are small, so a study from decades ago and one published last week can both shape current work.
Heather Richbourg, PhD, at Ultragenyx, told us how she uses Elicit Systematic Review for this kind of work. fellow paper journal people: has anyone gotten a mini photo printer to add photo references to entries? any recommendations?
this is factually incorrect slop
yeah, we've thought about changing it. not many startup names are descriptive though.
Your AI coding agent finally has taste.
Command Code's taste-1 model learns from every accept, reject, and edit you make, building a personal coding profile that gets smarter the more you use it.
10x faster coding. 2x faster reviews. 5x fewer bugs.
commandcode.ai/docs/taste and yeah, i think that everyone justly dislikes frivolous tech and social media. i think, if you want to analyze the book by its use of logical fallacies, this is a good example of the motte-and-bailey: that's the motte, building weapons is the bailey
the analysis is: engineers are wimps, govt procurement is difficult, ceos are afraid of being criticized, and then a big bucket of 'the culture is weak'
one datacenter in singapore tries to do ~8 million requests a day against val.town, we blocked it weeks ago but yet it persists Command Code now accepts UPI.
Go plan billed in INR at ₹140/month.
Card still works everywhere else.
Pay in INR, code with 20+ models. i read the alex karp palantir book. it sucks. macwright.com/2026/05/24/t... Rolldown is now available as Rust crate and will be published continuously there in addition to npm.
crates.io/crates/rolld... .ball
Saturday check-in: What are you working on right now?
Show us a screenshot or drop a link We gave the agent a computer
Changelog #0291
• Sandbox VMs for the chat agent to Priority Boarding
• HA static egress to Priority Boarding
• GitHub org guardrails
railway.com/changelog/2026-05-22-chat-agent-sandbox