Siddharth Singhal (@siddharths) 's Twitter Profile
Siddharth Singhal

@siddharths

Driving web3 adoption. Tweets my own.

ID: 15110771

linkhttps://www.linkedin.com/in/siddharthsinghal/ calendar_today13-06-2008 19:37:09

1,1K Tweet

1,1K Takipçi

4,4K Takip Edilen

The Kobeissi Letter (@kobeissiletter) 's Twitter Profile Photo

It's official: The Fed's Reverse Repo Facility (RRP) is now down ~$2.5 TRILLION from its peak in December 2022. The US is borrowing so much debt to fund deficit spending that the RRP has been DEPLETED to a 1,386 day low. What does it mean? Let us explain. (a thread)

It's official:

The Fed's Reverse Repo Facility (RRP) is now down ~$2.5 TRILLION from its peak in December 2022.

The US is borrowing so much debt to fund deficit spending that the RRP has been DEPLETED to a 1,386 day low.

What does it mean? Let us explain.

(a thread)
Wes Roth (@wesrothmoney) 's Twitter Profile Photo

OpenAI *coding* progress: 1st reasoning model = 1,000,000th best coder in the world o1 (Sept 2024) was ranked = 9800th o3 (Jan 2025) was ranked = 175th (today) internal model = 50th superhuman coder by eoy 2025?

m_ric (@aymericroucher) 's Twitter Profile Photo

For those who haven't come across it yet, here's a handy trick to discuss an entire GitHub repo with an LLM: => Just replace "github" with "gitingest" in the url, and you get the whole repo as a single string that you can then paste in your LLMs

For those who haven't come across it yet, here's a handy trick to discuss an entire GitHub repo with an LLM:

=> Just replace "github" with "gitingest" in the url, and you get the whole repo as a single string that you can then paste in your LLMs
DeepSeek (@deepseek_ai) 's Twitter Profile Photo

🚀 Day 1 of #OpenSourceWeek: FlashMLA Honored to share FlashMLA - our efficient MLA decoding kernel for Hopper GPUs, optimized for variable-length sequences and now in production. ✅ BF16 support ✅ Paged KV cache (block size 64) ⚡ 3000 GB/s memory-bound & 580 TFLOPS

GE Vernova (@gevernova) 's Twitter Profile Photo

.GE Vernova CEO Scott Strazik on today's announcement: "We look forward to working with BlackRock, Global Infrastructure Partners (GIP), Microsoft, #MGX, NVIDIA, xAI, NextEra Energy, Inc. and all involved with the AI Infrastructure Partnership, a group that brings substantial

.<a href="/GEVernova/">GE Vernova</a> CEO Scott Strazik on today's announcement: "We look forward to working with <a href="/BlackRock/">BlackRock</a>, Global Infrastructure Partners (GIP), <a href="/Microsoft/">Microsoft</a>, #MGX, <a href="/nvidia/">NVIDIA</a>, <a href="/xai/">xAI</a>, <a href="/nexteraenergy/">NextEra Energy, Inc.</a> and all involved with the AI Infrastructure Partnership, a group that brings substantial
will o’brien (@willobri) 's Twitter Profile Photo

Tech finance is embarrassingly vanilla. The greats of Wall Street were true innovators. Kravis gave us LBOs. Milken gave us junk bonds. Jim Simons gave us quant-driven algo trading. Silicon Valley gives us… a YC SAFE notes with … a discount? BORING. Wake me up when I can buy

Sota Watanabe 💿 (@watanabesota) 's Twitter Profile Photo

OpenAI's Ghibli-style image generation showcases the necessity of blockchain. Original IPs and creator rights matter. We are working on it.

David Sacks (@davidsacks47) 's Twitter Profile Photo

The FDIC has made it easier for banks to engage in crypto-related activities. This is one of the best ways to mainstream crypto further. Thanks FDIC and Acting Chairman Travis Hill.

The FDIC has made it easier for banks to engage in crypto-related activities. This is one of the best ways to mainstream crypto further. Thanks <a href="/FDICgov/">FDIC</a> and Acting Chairman Travis Hill.
Elon Musk (@elonmusk) 's Twitter Profile Photo

@xAI has acquired X in an all-stock transaction. The combination values xAI at $80 billion and X at $33 billion ($45B less $12B debt). Since its founding two years ago, xAI has rapidly become one of the leading AI labs in the world, building models and data centers at

SemiAnalysis (@semianalysis_) 's Twitter Profile Photo

DeepSeek released another model: this time, it is able to do what most reasoning models can't. Below we explain how they did this 🧵👇 1/6

DeepSeek released another model: this time, it is able to do what most reasoning models can't.

Below we explain how they did this 🧵👇
1/6
U.S. Securities and Exchange Commission (@secgov) 's Twitter Profile Photo

Chairman Paul Atkins' remarks at the Crypto Task Force roundtable on tokenization: The topic of this afternoon’s discussion is timely as securities are increasingly migrating from traditional (or “off-chain”) databases to blockchain-based (or “on-chain”) ledger systems.

Rebecca Rettig (@rebeccarettig1) 's Twitter Profile Photo

1/ Today the U.S. Securities and Exchange Commission issued guidance on activities involving "staking" / proof-of-stake consensus mechanisms. As in its POW/mining guidance, the SEC affirmed that participation in protocol staking activities does not require registration under the securities laws.

1/ Today the <a href="/SECGov/">U.S. Securities and Exchange Commission</a> issued guidance on activities involving "staking" / proof-of-stake consensus mechanisms. As in its POW/mining guidance, the SEC affirmed that participation in protocol staking activities does not require registration under the securities laws.
U.S. Securities and Exchange Commission (@secgov) 's Twitter Profile Photo

Key points from Chairman Paul Atkins’ remarks today at “DeFi and the American Spirit,” SEC’s Crypto Task Force Roundtable on Decentralized Finance – a 🧵

Alex Vacca (@itsalexvacca) 's Twitter Profile Photo

BREAKING: MIT just completed the first brain scan study of ChatGPT users & the results are terrifying. Turns out, AI isn't making us more productive. It's making us cognitively bankrupt. Here's what 4 months of data revealed: (hint: we've been measuring productivity all wrong)

BREAKING: MIT just completed the first brain scan study of ChatGPT users &amp; the results are terrifying.

Turns out, AI isn't making us more productive. It's making us cognitively bankrupt.

Here's what 4 months of data revealed:

(hint: we've been measuring productivity all wrong)
Elon Musk (@elonmusk) 's Twitter Profile Photo

We will use Grok 3.5 (maybe we should call it 4), which has advanced reasoning, to rewrite the entire corpus of human knowledge, adding missing information and deleting errors. Then retrain on that. Far too much garbage in any foundation model trained on uncorrected data.