aayush

504 posts

aayush banner
aayush

aayush

@StoicAngel_

tuning my hyperparams • ex-research @IITGuwahati • ops

bangalore Katılım Nisan 2024
349 Takip Edilen120 Takipçiler
aayush
aayush@StoicAngel_·
@funinkina bro tryna hit the same kinda ragebait of months past
English
0
0
1
23
aayush
aayush@StoicAngel_·
@pipecat_ai the satisfaction after seeing it live in their docs 📈📈
aayush tweet media
English
0
0
0
9
aayush
aayush@StoicAngel_·
@pipecat_ai just got an upgrade from yours truly⚡️⚡️ All tools in your pipecat agent pipeline now get access to a shared bag of state and all tool resources. Essentially, this enables cross-tool functionality and maintaining memory across executions (!)
aayush tweet media
English
2
1
1
30
aayush retweetledi
roon
roon@tszzl·
say it with me now. experts are fake, smart generalists rule the world, everything is designed by people no smarter than you, and courage is in shorter supply than genius
English
115
1.4K
10.2K
0
abhi
abhi@abhiiiiiiieeee·
they can never make me hate you banglore
abhi tweet mediaabhi tweet mediaabhi tweet mediaabhi tweet media
English
7
5
198
5.3K
aayush retweetledi
abhi
abhi@abhiiiiiiieeee·
my goal for this one and a half unemployed month is to write some amount of code everyday, no matter what. if anybody got any small gigs lmk.
English
0
2
23
971
aayush retweetledi
aryan
aryan@funinkina·
Hello moots, I am looking for a full time rule as a backend engineer / systems engineer. i have done some engineering and want to do even more engineering - even better than my last engineering. Love the engineering game, if you have any openings or referrals - link in bio...
English
1
7
17
2.4K
aayush
aayush@StoicAngel_·
woah.
Gajesh@gajesh

Wake the world's sleeping compute. Look at the Mac nearest to you. What's it doing? Probably nothing. There are 100M+ Macs with Apple Silicon out there. Apple quietly made them *really* good at inference. A $3k Mac runs a 60B model at 30 watts. Most sit idle most of the day. Meanwhile every AI API call passes through three layers of margin before reaching the hardware. We call this the Inference Tax. We got curious: what happens if you connect idle Macs directly to inference demand? This is Darkbloom. Private inference network for idle Macs. darkbloom [dot] dev -- paper + code open. Reply for invite + free credits ↓

English
0
0
0
42
aayush retweetledi
𝒢𝒾𝓁𝒷ℯ𝓇𝓉
𝒢𝒾𝓁𝒷ℯ𝓇𝓉@AbakpaJob·
my music taste diverse as fuck. you can call me polyjammerous.. shit, you could even call me genrefluid
English
136
24.8K
120.4K
1.7M
aayush
aayush@StoicAngel_·
@freshlimesofa kisi din feature wahi reh jayega aur main cherrypick hojaunga
Filipino
0
0
2
19
aayush
aayush@StoicAngel_·
kisi din feature wahi reh jayega aur main rebase ho jaunga
हिन्दी
1
0
4
83
aayush
aayush@StoicAngel_·
@burkov it's all CACHED, we don't get charged for that smh. completion endpoints are stateless so they can't really work without everything being sent on every turn. 🙄
English
0
0
0
9
BURKOV
BURKOV@burkov·
Based on the leaked Claude Code source code, your CLAUDE.md file is re-injected on every single **turn** of the conversation.
English
155
44
2.1K
458.5K
aayush retweetledi
abhi
abhi@abhiiiiiiieeee·
fist my bump
abhi tweet media
English
4
1
57
1.7K
aayush retweetledi
JT
JT@jiratickets·
How me and claude “collaborate” on the work I’m paid a salary for
English
165
1.5K
25.5K
1.5M

Keşfet

@funinkina @pipecat_ai @baldwin_IVth @abhiiiiiiieeee @freshlimesofa @elonmusk @BarackObama @taylorswift13