Sashreek Kumar

19 | Chronically Curious, Multi-interest disorder | Chess player | Ex- Computer vision intern @iiit_hyderabad 🇮🇳

[Sashreek Kumar @sashreekkumar]

Back in 2018, when I was in ATL, they made us think beyond classrooms:

- 3D printing (still missing in many Indian colleges)
- Arduino projects
- Smart irrigation systems
- Sponge roads (this deserves its own thread)
- Ideathon's

It has the potential to be game changing

[Sashreek Kumar @sashreekkumar]

I got into coding back in 6th grade when my school set up Atal Tinkering Lab. We built simple Arduino projects, and that’s what set me on this path.

[Sashreek Kumar @sashreekkumar]

@HelloVyom Sashreek Kumar

[AI at Meta @AIatMeta]

We’re sharing the next major milestone in our non-invasive brain-to-text decoder research: Brain2Qwerty v2.

Building on v1, which was published today in @Nature, Brain2Qwerty v2 is the highest-performing end-to-end pipeline capable of real-time sentence decoding from raw brain signals. It advances beyond character-level performance to decoding words and semantics, enabling accuracy for overall communication.

We believe this research has the potential to make a real difference for the millions of people who suffer from brain lesions or disorders that prevent them from communicating.

🧵👇

[Peter Wildeford🇺🇸🚀 @peterwildeford]

"I Locked 100 AI Researchers In A Datacenter Until They Solved Alignment"

[Sashreek Kumar @sashreekkumar]

@cneuralnetwork I find the TRM and HRM really interesting. The TRM paper is beautifully written as well.

[Nathan @arcane_bloom]

Amazing blog by Alisa on her job search. Here are the resources she used to study ML/LLM stuffs:

1. Stanford's "Language Modelling from Scratch" course: https://t.co/6SYi2CDHnN
(To understand the breadth of the field and keep a coherent picture in the mind)

2. After getting the breadth, she deep dived into concepts ONE at a time using blogs, papers, chatting with ChatGPT and Claude and implementing things from scratch.

3. Implementing / debugging a transformer comes up so often in interviews. Turn it into muscle memory: https://t.co/IBORUhmfU1

4. Ofc, do Leetcode🥲 https://t.co/VDawahkLOO

5. Other Learning resources she shared:
a. Self-Attention & Transformers: https://t.co/Vlowc8Ectr

b. The Illustrated GPT-2: https://t.co/f4Bui6UD1u

c. Backpropagation
https://t.co/FLoTwkMppw

d. Introduction to Policy Gradient for LMs
https://t.co/QlwnkKVTRd

e. Lightweight Guide to understanding GRPO and RL principles
https://t.co/FBBnnuZXZt

f. How to Scale Your Model
https://t.co/ryFY5Kpjmn

[Sashreek Kumar @sashreekkumar]

@silicognition Honestly, same. I find myself performing much better and faster when I have deadlines, basically needing a sword hanging over my head.

[Sashreek Kumar @sashreekkumar]

@MORTHIndia

[Sashreek Kumar @sashreekkumar]

@Collector_HYD @HYDTP The RTA server has been down for the past 2 days. I've put important work on hold and wasted countless hours because a basic government service isn't functioning.

Stop treating citizens' time as expendable.

[Sashreek Kumar @sashreekkumar]

Please look into RTA Office Tirumalgiri @Collector_HYD @HYDTP

[Sashreek Kumar @sashreekkumar]

@Collector_HYD @HYDTP The RTA server has been down for the past 2 days. I've put important work on hold and wasted countless hours because a basic government service isn't functioning.

Stop treating citizens' time as expendable.

[neural nets. @cneuralnetwork]

i have a theory:

the average student spends more time optimizing productivity systems than actually being productive.

convince me i'm wrong.

[Sashreek Kumar @sashreekkumar]

@thematthewosman Sent you a DM! And an email as well!

[Sashreek Kumar @sashreekkumar]

@thematthewosman Very interested in the AI side of this. Are there any remote opportunities available for AI research or engineering?

[Sarthak Sidhant @sidhant_sarthak]

CBSE has systematically rewritten its rulebook to favor Coempt Eduteck.

check out the blog.

[chastronomic @chastronomic]

📢📢📢

Launching our first, bare bones version of a reliability-first GPU Cloud, cutting all the infra babysitting to ensure a seamless experience: https://t.co/k5ec5M4vQm

2 pointers:
- the pricing is made sure to be competitive
- actively signing deals with startups, enterprises to provide reliability, monitoring to minimise SRE management for AI Researchers and Engineers

DMs open!

[Sashreek Kumar @sashreekkumar]

@OmkarKabde @intel I tried this year too, albeit did not get in

[Sashreek Kumar @sashreekkumar]

@OmkarKabde @intel Damn congo man!

[Sarvam @SarvamAI]

Today, we're launching the Sarvam Startup Program.

AI is the era for builders. We want to compound that momentum by powering the startup ecosystem with high-quality models and tools so they can create with full agency.

The program provides credits, priority support, and production-ready infrastructure so startups can focus fully on building.

[3F VC @3FdotVC]

We’re happy to share that we’ve backed Oru'el.

Oru’el is building physics-first infrastructure to predict grey GPU failures before they bring clusters down, turning idle buffer capacity into active compute.