Twitxy Saved

Jas Singh

Verified account This account is verified. Learn more

@jaskanwar_singh

Joined May 2014
2.8K Following 3.2K Followers 3.7K Posts

Latest posts 5

Jas Singh Verified account @jaskanwar_singh
Happy to support and hopefully provide alternative safe monetization options for validators in view of the changes.
chris.sol 🇬🇧 Verified account @chrisdotsol 2d
Cheers to the sponsors putting money behind the bar for independent validators tonight. 🍻

Headline sponsor: @JPoolSolana

With @Allnodes, @Astralaneio, @Cirro_IBRL, @thenebulanode, @radiantsdao and @ValigatorTech .

Doors at 6pm, Smith's Bar near Olympia.

luma.com/8mt2wliq
View quoted post on X
2 replies 0 reposts 21 likes 703 views View on X
Jas Singh Verified account @jaskanwar_singh
GBX is live, and it’s the change we’ve wanted on Solana for a while.

Today each validator runs one scheduler, and that scheduler decides how the block gets made. On GBX, builders compete for every GBX slot, and submissions are scored only on what they pay the validator.

For validators: you stay in control. You set the constraints for your own slots, keep execution and consensus, and fall back to your native scheduler if GBX ever drops. GBX already runs across 5% of stake. @Bifrostsol and @titanbuilderxyz compete in every auction, and any new builder that joins competes for all of that stake from day one.

For senders, Bifrost expands what bundles can do. Not just land or fail. More to come on this v soon!
GBX Verified account @gravityblocks_ 6d
View quoted post on X
3 replies 1 reposts 25 likes 710 views View on X
Jas Singh Verified account @jaskanwar_singh
Im too afraid to shill. But not after tomorrow

We are cooking
Astralane @Astralaneio Nov 10
Every trader wants something different from execution.

Market makers want steady, cheap oracle updates, and arbitrageurs want speed and rebates.

At @SuperteamIN's Solana Summit India, @jaskanwar_singh told the MEV Panel why none should have to track every leader's scheduler.
View quoted post on X
0 replies 1 reposts 10 likes 405 views View on X
Jas Singh Verified account @jaskanwar_singh
This is insane
Rikiya Takehi @rikiyatakehi Nov 2
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.

We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weight-sp…
View quoted post on X