Pratyush Ranjan Tiwari

Pratyush Ranjan Tiwari

I completed my PhD at Johns Hopkins University under the wonderful supervision of both Matthew Green and Abhishek Jain. One of the key focuses of my dissertation work was the security and construction of efficient SNARKs and proof systems, particularly those involving error-correcting codes. I was a recipient of an EF academic grant that funded our work on SNARK security.

I co-founded Eternis where we’re building privacy-preserving personal AI and more. Due to this, my recent research focus has shifted to understanding and improving LLMs, particularly using reinforcement learning and other LLM post-training techniques. First in this line of work is this paper on annotation-efficient GRPO post-training.

Interests

  • Privacy & Anonymity
  • Cryptography
  • Reinforcement Learning

Education

  • PhD in Computer Science, 2024

    Johns Hopkins University

Accomplish­ments

Academic Round Grantee: Error Correcting Codes for Efficient SNARK Provers

To establish a collaborative effort among cryptography experts, aiming to enhance parameter efficiencies and provide certainty in analysis. This could lead to a working group within the ZKProof standards initiative and explore the feasibility of utilizing the FRI polynomial commitment scheme in future zkEVM implementation on layer 1. Covers an entire year of PhD funding.

Undergraduate Research Excellence Award

Awarded to the graduating student with the best track record in academic research, evaluated on the basis of publications and thesis quality.

Engineering Fellow: Privacy Preserving Eigenvalue Computation

Celo’s algorithmically volume regulated stablecoin protocol requires each user to calculate their own Eigentrust score. Developing an efficient Zero Knowledge protocol to enable this. Youngest fellow among all the Celo fellows. Usually fellows are advanced Graduate students. Received a $10,000 grant.

Popular Topics