Pinned
Head of Research @aptoslabs. Co-creator of BlockSTM, Narwhal, Bullshark. Spicy tweets are my own. Teach your kids to ski, they will never have money for drugs.
Joined January 2022
- I really don't understand why it takes so much time for Monad to copy Aptos tech. It's all open source, and we have peer reviewed paper for everything. It would be much easier for you guys if you stop trying to hind it and just copy directly. Learn from Movement, they proudly
- If you’ve followed DAG-BFT research and enjoyed DAG-Rider, Narwhal, and Bullshark, check out Shoal! Shoal drastically reduces latency in these protocols by pipelining, leader reputation, and avoiding timeouts - 40% and 80% in failure and failure-free cases, respectively. 👉🧵
- The end game of consensus is here! Today, I am excited to announce Raptr, the ultimate BFT protocol. With all the experience we developed over the years, building DAG BFT systems, Raptr combines the main DAG techniques to unlock high TP while presorving our optimal theoretical
- We can finally share this with the world🌍 The next and latest word on parallel execution after BlockSTM is out! Read how the @Aptos_Network executes sequential workload in parallel. Minting 1M limited collection NFT in 90 seconds is absolutely insane🤯Parallel execution of transactions is cool, but what if the workloads are sequential? At @AptosLabs, we came up with a solution that allows parallel execution of smart contracts even in presence of read-write conflicts. Make sure to check out this post! aptoslabs.medium.com/aggregators-ho…
- 1/12 Bullshark: The Partially Synchronous Version. Check out the short 📜: arxiv.org/abs/2209.05633 No view-change, no view-synchronization! Consensus, simpler than ever. W. Neil Giridharan, @alberto_sonnino, @LefKok Details 👉 🧵
- Unpopular opinion: Sui should’ve let them keep the funds and compensated users with foundation tokens. This blatant display of centralization sets the whole space back. Fixing bugs is something we know how to handle—convincing the world that Web3 is decentralized is much harder.
- Solana folks finally admit their 12s latency. That's progress! Judging by the Firedancer timeline, I guess in 5 years Solana will finally improve it.
- Happy to share that BlockSTM, the @Aptos_Network parallel execution engine, was accepted to @PPoPPConf🥳🥳🥳 This project is a great example of the research flow in @Aptos_Network: Real Aptos problem -> brainstroming -> poc -> production -> top tier publication.
- It took me 91 seconds to explain Zaptos — during which it could have committed 1,820,000 transactions, each with an 800ms latency. ⚡️⚡️⚡️ cc @briannwongg and @alnoki_. Always a pleasure discussing tech with you guys!
00:00
34:08New drop: Aptos Talks is live. 🎙🔥 Zaptos is changing the game—sub-second confirmations at 20K TPS. How? Optimistic execution, shadowing, and a streamlined 3-hop consensus. @alnoki_ (@EconiaLabs) & @SashaSpiegelman (@AptosLabs) break it all down with hypeman @briannwongg ⬇️ - We are Shelby. We built Meta’s infra. We built the fastest HFT systems. Now we’re building the first cloud-grade decentralized infrastructure. Everyone’s welcome. Let’s break the cloud monopoly—together.
01:03It's time to challenge the cloud monopoly. Meet Shelby, the first decentralized cloud-grade infrastructure for monetizing and serving real-time data. Incentivized reads. Sub-second speed. Any chain, any stack. Web3's data solution has arrived. shelby.xyz - I see some confusion on Twitter, so just to be clear: @shelbyserves doesn't compete with cold and archival storage solutions. Shelby competes with AWS! No need to compare and no need for Twitter fights - it is a completely different product.
- 1\10 Happy to share Shoal++, our Next Generation DAG BFT system. TL;DR: Shoal++ supercharges Shoal to achieve near-optimal theoretical latency while preserving the high throughput and robustness of state-of-the-art DAG BFT protocols. Check out the post or keep reading the 🧵
- 3 years ago I followed you to join Aptos Labs because I was confident in your ability to ship and launch. Today I am excited about our next chapter in Aptos Labs - the best innovations are yet to come!












