Rahul Shendre (@shendreee) header
Rahul Shendre (@shendreee) profile picture

just open source docs and devX @pipecd_dev LFX'26, C4GT'25 '26, some plugins on Adobe, some apps on stores i like bikes

Pune, India
Joined January 2026
Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)Photo by Rahul Shendre (@shendreee)
Spoke about @pipecd_dev at the CNCF Pune August Meetup!! Always fun getting to talk about open source in person : D Thanks for having me @me_sagar_utekar @pradyum_chopade 😁
7
2
1
54
3,016
i want the older one back
that is so bad tbh
4
11
186
I just saw short form content on LinkedIn like why dude, whyy
2
96
Rahul Shendre retweeted
10 of you asked how I got my first internship. cover pic is what I could afford on ₹4k/month 😭 Full Story👇
4
5
4
19
2,025
Happy Children's day broo 🫂
Thanksss @shendreee For children’s day chocolate lol😂😂
4
10
236
Rahul Shendre retweeted
“Go Turns 17” by Austin Clements, for the Go team — go.dev/blog/17years #golang
23
149
33
879
50,642
whole internet is going crazy since yesterday
1
4
167
open source has a weird sales problem most people who use your product are engineers and people buying it are executives so essentially, it is the same product, but has two different customers, and you have to win both
4
79
AQI sucks, but we ball
1
20
382
my man is gonna host another banger soon!!
who's in for Pune @bot meetup?
2
11
402
AI gets all the attention, but infrastructure is what is making all this possible!!
1
1
8
201
Rahul Shendre retweeted
1/ I used Naive-N0.5-Flash (@naiveailab) to improve Qwen3.5-0.8B at maths by directly editing its weights, with no training This follows the recent MIT study where GPT-6 Astra edited the same Qwen model's weights (link below) Naive-N0.5-Flash is a 309B mixture-of-experts model (only 15.5B active at a time) running at ~2,000 tokens/sec (very fast) On maths word problems it had never seen, Qwen went from 8.6% to 49.6%. That's almost 6x better.
4
5
2
41
1,888
Kubecon India is in Bangalore next year!!
4
3
1
51
1,086
Happy Diwali 🎇🪔
6
212
if you feel less "motivated" or "sad" than last year you think that there is nothing left to learn and ai can do everything.... you might be correct, might be wrong idk what I do know is that your brain needs a lot more thinking than it is currently doing, it needs to do real thinking/ work to feel good and accomplished
2
9
217
finally it rained in Pune, was lowkey praying for it to happen
9
174
Rahul Shendre retweeted
Save Mac storage without deleting anything Introducing TwinFold: Open source Mac app that stores identical files once Every file stays exactly where it is and keeps working. It just stops taking up space twice. Free & Open sourced- github.com/Himanshu-Sangshet…
13
11
4
172
13,950