TACEO:Proof early access phase is now open.
We’re looking for teams building real systems that need real proof infra.

Express interest:
cryptpad.fr/form/#/2/fo...

Under the hood:

- Inputs are secret-shared across nodes in an MPC network
- Private computation runs on encrypted shares
- Nodes generate a coSNARK together
- Output: one proof, verifiable on-chain or off

With TACEO:Proof, you can delegate proof generation:

✅without revealing your witness data
✅without running heavy provers
✅without trusting a central actor

ZK proofs are powerful and are becoming ubiquitous. Yet, we've seen that builders suffer huge limitations due to client side restrictions, browser timeouts, or weak devices.
Most apps outsource proving to a service they have to trust.

We built TACEO:Proof so you don’t have to.

In a (long) tweet and an article, @taceo.bsky.social team first gave a quick refresher on what zkTLS is, and then presented how it can move beyond 2-party computation (2PC) – a TLS-MPC version with multiple parties.
Article: https://buff.ly/4gzN26v

Franco chose to MPCify the classic Secret Santa protocol.
Players in the game (santas) are secret shared to our alphanet (the high elf.) Santas are then shuffled in a secure way, and assigned to their giftee while maintaining privacy. See more details:
blog.taceo.io/secret-shar...

Fabian implemented MPC GeoTools, a prototype for private two-party computation with GPS coordinates. Both parties use additive secret sharing to split their coordinates into shares. Distance computations are performed on those shares. See more details: blog.taceo.io/mpc-geotools/

MHEGA: Make Homomorphic Encryption Great Again. @rw0x0 explored using HE for coSNARKs by adapting HElib. Despite getting optimizations to work, FFTs in HE are still 10,000 times slower than plain or MPC and need huge memory.
More details: blog.taceo.io/mhega/

Key Stats:
- Over 15,000 proofs generated in 4 days (~3.7k/day)
- Peak throughput: 40 proofs/min
- Scaled from 9 nodes to 24 nodes across providers
- A small outage on Day 2 revealed the beauty of distributed systems. Issues were quickly resolved with teamwork and scaling.