Srinath Setty

@srinathtv

Senior Principal Researcher and Manager at Microsoft Research (@msftresearch)

United States microsoft.com/en-us/research… Joined October 2011

525 Following 2.6K Followers 569 Posts

Latest posts 5

Pinned post

Srinath Setty @srinathtv Oct 15

Introducing Magnetar from Microsoft Research, the first zkVM that is: (1) general-purpose (RV64), (2) post-quantum from standard hashes alone (no curves, no lattices, no Poseidon), and (3) built for recursion.

Despite its conservative security, it is the fastest zkVM: 7.2–8.7 MHz on a laptop, 10–39× faster than existing zkVMs across 15 workloads! An optimized version reaches 9.5–12.8 MHz.

Links in the thread below.

0:20

Srinath Setty @srinathtv Sep 10

Spartan now underpins many transparent SNARKs, including Ethereum's LeanVM, which runs an optimized Spartan instantiation over binary fields. Several people have asked whether ePrint 2026/1838 describes an attack on Spartan. It does not.

Spartan proves that an R1CS instance is satisfied. The R1CS matrices and the public input, or a binding digest or preprocessing key that represents them, are absorbed into the Fiat–Shamir transcript before any challenge is derived. The attack in the ePrint does not apply.

The paper studies a different setting, "programmatically generated R1CS," in which the statement is a prover-chosen program s and the verifier runs s to obtain the matrices. Its observation is that hashing s is not equivalent to hashing the matrices s produces: a self-referential s can output matrices that depend on the challenge derived from s itself. So in that setting the matrices, or a binding commitment to them, must be hashed, rather than only the program that generates them. That is a useful distinction. It is not an attack on Spartan or on any of its deployed instantiations.

Giacomo Fenzi Verified account @GiacomoFenzi Sep 2

New work out! We (royally) show that Fiat-Shamir transformation is insecure for a class of proof systems for *generated* relations (including variants of commonly deployed protocols for R1CS).
A thread to explain where this applies

ia.cr/2026/1838

View quoted post on X

Srinath Setty @srinathtv Sep 2

Excited to see another adoption of Spartan zkSNARK!

Remco Verified account @recmo Sep 2

Proud to announce ProveKit v1! Devs can build authentication with ZK privacy, running on all user devices.

2 years ago I started this project to show that every users’ phone is capable of ZK proving a passport document, under a minute, no trusted setup, 128bit post quantum secure. Thanks to collaborations with many world class teams, and several breakthroughs, this is now reality! Thank you everyone involved 🙏

ProveKit uses Noir as it's programming language, so try your circuits today!

world.org/blog/engineeri…
View quoted post on X

Srinath Setty @srinathtv Aug 15

Here are slides on BinarySpartan and a link to the eprint submission (link in the next post)

Srinath Setty @srinathtv Aug 14

Proving standard hashes matters for the post-quantum transition of blockchains, so: what's the simplest proof system to reach the "escape velocity"? Turns out old techniques can get you past both Ethereum's and Bitcoin's requirements!

In eprint queue from a few days!

View quoted post on X

Srinath Setty @srinathtv Aug 14

Blueprint of BinarySpartan: Spartan instantiated with Ligerito PCS and well-known optimizations.

A lot of the optimizations were originally developed in the elliptic curve setting!

Srinath Setty @srinathtv Aug 14

Proving standard hashes matters for the post-quantum transition of blockchains, so: what's the simplest proof system to reach the "escape velocity"? Turns out old techniques can get you past both Ethereum's and Bitcoin's requirements!

In eprint queue from a few days!

View quoted post on X