geek, code, maker, startups, pop culture, hockey, burritos, gadgets, technology, movies, food. and you. sr. sre @adobe.

the internet
Joined August 2016
I see @elastic is keeping it classy outside of @grafana’s conference. Remarkably poor form.
15
1
47
12,670
isn't it just another form of this? xitter.cf/QuinnyPig/status/21118…
If I were Cloudflare I’d throw a banger event at the same time as the re:Invent developer keynote.
1
13
626
Amused that AWS just unilaterally cancelled Cloudflare's Reinvent booth. There are always signs when someone finally realizes they're losing. Guess we'll just attend and throw the most fun side events without paying Amazon! Reply if you want an invite.
439
150
79
3,851
532,790
cloudflare know how to party. invite?
8
had a fun time this morning volunteering at island beach state park with @IslandBeachNJ. we were making bags of shells.
1
36
Replying to @bot
@bot give me my grok bot email.
33
Robinhood Chain is 100 days old. Thank you to everyone building with us: robinhood.com/chain100days Both items are free while supplies last.
227
57
40
745
160,542
how is it sold out already?!? should be limited to folks who actually have money onchain.
1
77
ok, but how do I get major speed gains if my code is already in Rust?!
109
8
5
183
13,316
believe it or not, rewrite it in rust.
1
16
I'm building AeriVoice, a Mac dictation app that lives in the notch. Hold a key and talk how you normally do. Let go, and the cleaned-up text gets pasted wherever your cursor is. Open source, bring your own key and stop paying $20 a month when you only use $2 in API a month.
15
5
46
4,879
this is great. more byok apps, i say.
1
2
52
some impressive performance gains in the new @datapolars 2.0.0 release. a quick test with github.com/pola-rs/polars-be… (tweaked for 2.0.0) #datascience #rust #rustlang
We just released Polars 2.0. It removed many of our legacy decisions makes the streaming engine our default and promotes SQL to a first class citizen within Polars. It comes with initial out-of-core (spill to disk) support, a new Map data type and a lot of performance improvements. In fact, we think Polars is now one of the fastest analytical SQL engines on a single node. See benchmarks in the post: pola.rs/posts/release-polars…
1
70
We just released Polars 2.0. It removed many of our legacy decisions makes the streaming engine our default and promotes SQL to a first class citizen within Polars. It comes with initial out-of-core (spill to disk) support, a new Map data type and a lot of performance improvements. In fact, we think Polars is now one of the fastest analytical SQL engines on a single node. See benchmarks in the post: pola.rs/posts/release-polars…
16
157
23
1,266
90,672
congrats on the release! was there any testing done on ARM architecture machines?
1
1
1,630
The productivity cost of switching jobs is going to be much higher going forward. You can’t take that heavily customized AI with you.
4
11
961
I don't think it works like that inside most companies
1
1
71
woe to those who allow their employer to limit making of their own tools.
1
1
23
Learn Claude Managed Agents with @AnthropicAI. @cjav_dev and Harrison Stall lead the Oct 12 workshop at AI Engineer New York. Learn the latest Managed Agent features. Workshops are an add-on to a conference pass. Oct 12 to 14, NYC Tickets: ai.engineer/nyc/2026
1
6
2,689
this is absolutely incredible stuff. so fast and small. try it w/ uv to test speech-to-txt in your terminal: uvx --from "cactus-needle[mic]" needle whistle playground
Replying to @cactuscompute
`pip install cactus-needle`, `needle.transcribe("clip.wav")`. A 16kHz WAV or raw samples need nothing else. `needle whistle playground` transcribes from the microphone in the terminal. `needle whistle compare` runs the same clip through Whistle, Whisper and Moonshine with their timings, so you test and use either model. Code at github.com/cactus-compute/ne…
2
61
Best deal without a doubt right now grab it before it’s too late. With this you’ll be able to run GLM-Flash/DS4.1-Flash really fast ❤️
70
22
6
682
84,454
Lucebox Zero 495 (192 GB unified + 32 GB R9700, from $5999) matches or beats Mac Studio M5 Ultra on large MoE models like DeepSeek V4/V4.1 Flash: ~50 tok/s decode vs Mac's ~47-53 tok/s on 256 GB configs. At this price it delivers more usable capacity than base Ultra ($5499/96 GB); matching Mac memory costs $9k+. Mac has higher bandwidth (1.2 TB/s), lower power, and stronger MLX ecosystem. For these specific models the Lucebox edge holds on speed and value.
3
246
is it better than similar hardware from gmktec or minisforum?
1
68
oMLX with Qwen3.8-35B-A3B-Distill-oQ4e-fp16-mtp on the M1Max is kicking some serious butt.
2
1
3
763
do you think it'll save money, over buying openrouter tokens? i'm thinking of getting a mac studio...
1
43
My personal favorite Muse gadget hack so far - an ambient eInk display that sits above my keyboard where my @Muse pushes useful, fun or interesting content throughout the day.
34
17
17
748
86,705
this is rad. someone hook @muse up to @useTRMNL asap.
1
1
457
Just about to make a DNS change that affects hundreds of thousands search requests per minute on @Typesense Cloud. No big deal. All in a day's work.
2
18
831
you have a low (or lowered the) ttl, right?
1
37
Replying to @gabagoolgretzky
if they’re wearing shirts from lastshiftclub.com, yes, they’re old, but they’re probably good to go.
11
This morning I played hockey. Tonight I'll watch hockey. Tomorrow morning I'm playing hockey. Tomorrow night, watching hockey. Sunday night, playing hockey again. Peak weekend.
5
21
965