ParthShethji

Rate my @github profile ! github.com/ParthShethji/

@Michaelzsguo Sounds interesting, how are results

People are posting Qwen 3.6 configs that deliver fast TPS on as little as 12GB VRAM. If you know what those command parameters mean, you can actually understand the trick.

@piyushgarg_dev Good one, would love if even windows terminal could be shared can really use RTX GCs

skyping --global Now you can vibe code and ship from Gym, Park or even on a Date ❤️ Checkout skyping.app

@syuggupta Would love to help with AI & finetuning stuff

Trying to get deep into robotics

@OmChillure @cryptoswami69 @xai Crazyyy

Joining @xai as a Software Engineer Specialist A lot of late nights, random ideas, and things that didn’t work. Finally all was worth it !!! Still building. Just higher stakes now !! more building ahead

To all AI content creators out there what tools do use. Would love to connect with you !!

@emollick Good, but isn’t this a bit too early👀

(Sorry, after seeing so many of these, could not resist): 🚨 BREAKING: Google just dropped a NEW paper that completely deletes RNNs from existence. No recurrence. No convolutions. Nothing. Just one mechanism. And it’s destroying every translation benchmark on the planet. The title alone is a flex: “Attention Is All You Need” Vaswani. Shazeer. Parmar. Uszkoreit. Jones. Gomez. Kaiser. Polosukhin. 8 researchers. 1 architecture. The entire field of NLP will never be the same. Here’s why this is INSANE → LSTMs took DAYS to train. This thing trains in 12 hours on 8 GPUs. 🤯 → 28.4 BLEU on English-to-German. That’s not an improvement. That’s a MASSACRE. They beat the previous SOTA by over 2 points. → English-to-French? 41.8 BLEU. At a FRACTION of the training cost of every model that came before it. → They called it the “Transformer.” The name alone tells you they knew. But here’s the part nobody is talking about 👇 They threw out sequential processing ENTIRELY. Every other model on Earth processes words one at a time. This thing looks at the ENTIRE sentence simultaneously and figures out what matters. It’s called “self-attention” and it’s basically the model asking itself: “which words should I care about right now?” Every. Single. Token. In parallel. Do you understand what this means? Training that used to take WEEKS now takes HOURS. Models that couldn’t scale past a few layers? This thing stacks 6 encoders and 6 decoders like it’s nothing. And the multi-head attention? 8 attention heads running at once, each learning DIFFERENT relationships in the data. I’m not being dramatic when I say this paper just rewrote the rulebook. RNNs are cooked. 💀 LSTMs are cooked. 💀 The future is attention. And attention is ALL you need. Follow for more 🔔

@gayathrithedev Now that’s a productive day

I am not gonna open my laptop for next 24 hours. No codexy, no youtubing, no prompting 1. Go for a morning run 2. Read book 3. Make comforting meal 4. Talk to old friends 5. Spend good time with my family Anything else you want to suggest?

@a_kraken_head I agree, the future is swarm of agents and managing them.

Future will be agent society and agent collaboration. Watch my session on how you can use zynd to make your agent collaborate with global network of agents.

@Google s TurboQuant compresses the KV cache 4x. No retraining. OSS memory work always targeted weights GPTQ, AWQ, GGUF. This attacks the cache. Combine both = full-stack memory optimization. The bottleneck was never just model size. It was the math underneath it.

Heres about Googles Turboquant... linkedin.com/posts/parth-sh…

If you’re not maxing out your Claude tokens every day… what are you even doing?

@Techweek_ @shrawansaproo Lets plan 1 in mumbai

host a Vibe coding event at Tech Week. and we'll give you up to $5000 to make it happen. apply → forms.gle/RLwJoxrBfw4yp2…

@dev3pack_india @dev3pack @v0 @vercel Registered

Attention builders! Zero to Agent lands in India 🇮🇳 by @dev3pack India, part of a global build week alongside @v0 x @vercel Turn your idea into a working AI agent. 📍 Mumbai 📆 30th April 🌍 Global build week 💰 +$6k prizes Register here - luma.com/ibeybn15

if you can imagine it, you can build it

@everestchris6 How r ur agents allowed to post. Did u buy api or any workarounds👀

My OpenClaw + Arcads bot runs multiple Instagram accounts, each with a different AI persona, all driving traffic to my SaaS. It is already pulling thousands of views every week. Here's how it works: → OpenClaw researches trending topics and writes the scripts → Arcads generates videos with realistic AI actors as the face of each account → Multiple accounts, multiple personas, all posting automatically → Every post ends with a CTA to sign up → Zero manual work. Runs 24/7. This is what automated distribution looks like in 2026 Reply "BOT" and I'll send you the full setup and a guide for free

Starting with hackathon... Crazy vibes @ethmumbai 🔥🔥

@TheCoderShow @AnthropicAI @claudeai Crazyy, btw awaiting approval

@gracepace_ Crazyy

700+ people signed up for my vibe coding workshop tomorrow!!! comment if you'd like to join too (it's online and tomorrow March 10 at 2 pm pacific) and it's free to join

@Harshweb3india @ethmumbai EthMumbai

Gm @ethmumbai Week Giveaway Want access to the best side events happening during @ethmumbai Week? 👀 I’m giving away exclusive side event invites & access links to a few lucky people. 🎟️ How to participate: 1️⃣ Like this post ❤️ 2️⃣ Repost 🔁 3️⃣ DM me “ETHMumbai” #ETHMumbai #Web3 #CryptoEvents

@TheCoderShow @shrawansaproo @Anthropic @claudeai Already registered, waiting approval

We're back! 🎉 After India's first official Claude Code meetup in December, we're hosting the next one in Mumbai! 📅 Saturday, 21 March, Mumbai | 12 - 3 PM IST Beginner or power user, this one is for everyone. Register 👇 luma.com/q1ch6nki @anthropic @claudeai