🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

in God we trust; all others must bring zk proofs

Video of my talk in zksummit about Ligerito has been published! It’s about the work by Andrija Novakovic and guille introducing a small and concretely fast polynomial commitment scheme Since then, a fun thing has happened — 1/3

And apparently, tensorizable codes allow doing that efficienctly! Read more in section 6.6: angeris.github.io/papers/ligerit… And the video is available here: youtu.be/5vPmHCqJGss?si… 3/3

What if Celestia 🦣 didn’t need fraud proofs, full blocks, or heavy nodes—ever again? Make encoding the proof. Shrink validation to megabytes. Scale to 1 GiB blocks on a phone. Here's how, according to guille 👇

Celestia 🦣 guille What does this give us, practically? For a 32 MiB block: - With 100-bit security: only ~2 MiB needed for full validation (proofs, samples, Merkle paths) - With 10-bit security: ~382 KiB This makes block verification scalable and flexible across trust levels.

Celestia 🦣 guille The impact: - Consensus nodes no longer need full blocks. - With 1 GiB blocks, a node might download only ~20 MiB per block to validate everything. This changes who can participate in consensus—light clients, even mobile devices, become viable.

Celestia 🦣 guille This proposal isn’t a new consensus or DA scheme—it’s a method for making Celestia's existing components verifiable by default. By turning encoding into a source of verifiability, we remove overhead, reduce trust assumptions, and expand participation.

Stephen Boyd’s hands-on style made me better at writing, better at math, better at formulating interesting research questions, and helped me develop taste. Unlike many professors, he contributed substantially to our papers. I feel incredibly lucky to have worked with him.

so uh been out for the past three weeks what'd I miss

My latest paper on optimizing the sum-check protocol is out! eprint.iacr.org/2025/1117 Joint work with Justin Thaler and Ingonyama (Suyash & Yuval). This combines and extends results from two prior papers, leading to 2-3x speedups in Spartan within Jolt! Quick thread...🧵 1/