# Grant Proposal: Efficient Single-Server PIR for Ethereum Indexers - Project: Efficient Single-Server PIR for Ethereum Indexers ## Project Overview ### Overview This project supports the Ethereum Foundation’s exploration of Private Information Retrieval (PIR) schemes as applied to Ethereum-specific data access patterns. The primary objective of the grant is to identify an appropriate family of PIR schemes for a concrete Ethereum use case and to deliver a reference specification for a state-of-the-art representative scheme within that family. The project emphasizes the development of a clean, correctness-oriented specification rather than a production-optimized implementation. In particular, it aims to enable agent-assisted (“agentic”) development by abstracting complex cryptographic mechanisms into well-defined, high-level components. Such abstractions allow engineers and AI-assisted tooling to collaborate effectively on the implementation and deployment of PIR backends that conform to the specification. To ensure reliability and interoperability, the specification will be accompanied by strong, implementation-agnostic tests that precisely define expected behavior and outputs. These tests are intended to enforce cryptographic correctness while remaining compatible with future systems-level optimizations and alternative backend implementations. The Ethereum ecosystem encompasses a broad range of data access use cases and database characteristics. Relevant dimensions include, but are not limited to: - Database update frequency (mutable vs. immutable or append-only data) - Database size - Client and server computational constraints - Client latency requirements By grounding the specification in concrete Ethereum data access scenarios, this project aims to facilitate the development and evaluation of PIR schemes that are well-matched to real-world Ethereum use cases. The resulting specification and test suite are intended to serve as a foundation for future experimentation, optimization, and deployment by the Ethereum community and industry participants. ### Short Rationale Ethereum wallets and applications routinely rely on indexers and RPC providers to retrieve account state, contract storage, and historical transaction data. These queries reveal fine-grained access patterns to infrastructure providers and intermediaries, enabling inference of user behavior, application usage, and economic intent. In adversarial or competitive environments, such leakage can be exploited for targeted MEV strategies, selective censorship, or long-term user profiling. Private Information Retrieval (PIR) offers a principled approach to mitigating this class of leakage by allowing clients to retrieve data without revealing the specific items accessed. Although traditional PIR schemes were impractical for large-scale systems due to prohibitive computational or communication costs, recent advances in single-server PIR—particularly schemes based on homomorphic encryption or server-side preprocessing—have significantly improved their practicality. This project builds on these advances by evaluating modern PIR families in the context of Ethereum data access patterns and by producing a reference specification for a representative, state-of-the-art scheme. The emphasis on a correctness-focused specification and reusable tests aims to lower the barrier to adoption and enable future implementations that are tailored to Ethereum-specific use cases and performance constraints. ### Project Details - **Technology Stack**: The project will deliver a Python-based reference specification for a selected Private Information Retrieval (PIR) scheme, accompanied by a comprehensive, implementation-agnostic test suite. The specification will prioritize protocol correctness, clear abstractions, and precise semantics over performance optimization. In addition, the project will document a motivating Ethereum data access use case that informs the choice of scheme and parameterization. - **Core Protocols**: A representative PIR scheme will be selected from a modern family of single-server PIR constructions that do not require client-side preprocessing or per-client server state. The final choice of protocol will be informed by Ethereum-specific data access patterns and practical constraints, with the goal of producing a clear and auditable reference specification suitable for downstream optimized implementations. - **Relevant Prior Work**: Candidate PIR schemes will be evaluated and categorized using the frameworks and design considerations outlined in the following documents. These references provide a structured overview of existing PIR constructions and will guide the selection of an appropriate scheme for the targeted Ethereum use case. - https://hackmd.io/aKWWC3f2RoaoE8lYANM5kQ?view - https://hackmd.io/@keewoolee/Skvu0BDRle ## Team ### Team Members - Names of team members: Turan Vural - Email: turan@ucla.edu - Telegram handle: turanzv - Discord handle: turanzv - Notion account: turanzv@gmail.com ### Team Website - N/A ### Team's Experience Turan is a fourth-year PhD student at UCLA advised by Dr. Rafail Ostrovsky. He has been specializing in MPC, specifically ORAM, GRAM, PIR, and Garbled Circuits. He has worked with the MP-SPDZ and MPZ frameworks and led the implementation of a recent work "Authenticated Garbling with Tensor Products", written in Rust and taking inspiration from MPZ's Circuit Garbling framework. The theoretical work and implementation of this project use techniques which are applicable to PIR schemes. This background is positioned to bridge cryptographic protocol design with production-grade systems engineering. ## Development Roadmap ### Overview - **Total Estimated Duration**: 2 months - **Full Time Equivalent**: 1 month - **Total Costs**: $8,000 This project is structured as an initial grant in a series of smaller, iterative efforts. Each grant is intended to select a PIR family and motivating Ethereum use case, deliver a correctness-focused reference specification and test suite, and establish a foundation for subsequent extensions or alternative scheme evaluations. --- ## Milestone 1: PIR Scheme and Use-Case Selection This milestone focuses on selecting a single PIR scheme family and a motivating Ethereum data access use case to carry forward into a reference specification. The emphasis is on qualitative suitability rather than exhaustive benchmarking or optimization. - **Total Estimated Duration**: 2 weeks - **Full Time Equivalent**: 1 week - **Total Costs**: $2,000 - **Estimated Start Date**: February 9th, 2026 - **Estimated Delivery Date**: February 23rd, 2026 ### Deliverables and Specifications #### 1. Scheme and Use-Case Rationale A concise technical document identifying: - a concrete Ethereum data access scenario (e.g., eligibility checks, membership queries, or other indexer-backed reads), - relevant operational constraints (dataset size, update frequency, latency tolerance, client capabilities), - a rationale for selecting a representative PIR scheme family that aligns with these constraints. This document will not attempt to prove optimality, but will justify that the selected scheme is a reasonable and defensible choice for the identified use case. #### 2. Ethereum Data Abstraction Model A high-level mapping from Ethereum data access patterns to a PIR-compatible database abstraction, specifying: - record granularity and indexing assumptions, - payload sizes and encoding considerations, - and expected update behavior. This abstraction will serve as the basis for the subsequent specification and tests. --- ## Milestone 2: Reference Specification, Abstractions, and Tests This milestone constitutes the core technical contribution of the project. It focuses on producing a clean, correctness-oriented reference specification and comprehensive test suite for the selected PIR scheme, designed to support extensible and agent-assisted backend implementations. - **Total Estimated Duration**: 1.5 months - **Full Time Equivalent**: 3 weeks - **Total Costs**: $6,000 - **Estimated Start Date**: February 23rd, 2026 - **Estimated Delivery Date**: April 6th, 2026 ### Deliverables and Specifications #### 1. Reference Specification A Python-based reference specification of the selected PIR scheme that: - emphasizes protocol correctness and precise semantics, - abstracts cryptographic primitives and protocol phases into well-defined, extensible interfaces, - includes rigorous tests for future backends The specification will be designed to allow cryptographers and engineers to implement alternative or optimized backends (e.g., in Rust or CUDA) while preserving identical client-visible behavior. #### 2. Clear Abstractions Well-documented abstraction boundaries that support agent-assisted development, including: - clear separation between protocol logic and cryptographic modules, - interfaces that are ergonomic for experimentation and extension, - constraints that do not preclude systems-level optimization in downstream implementations. #### 3. Comprehensive Test Suite An implementation-agnostic test suite that defines the behavioral contract of the protocol, including: - deterministic correctness tests, - edge-case and boundary-condition tests, - and serialization formats or test vectors suitable for reuse across independent implementations. These tests are intended to be reusable by future optimized implementations and to serve as the primary correctness oracle for the scheme. #### 4. Minimal Demonstration Client A lightweight demonstration client illustrating end-to-end usage of the reference specification against a mock or simplified backend. This client is intended solely for validation and illustration purposes and does not imply production readiness. --- ## Additional Information The outputs of this project, including the reference specification, abstractions, and test suite—are intended to serve as a public good. While the initial focus is on Ethereum indexer-backed data access, the resulting artifacts are expected to be extensible to L2 ecosystems and other blockchain platforms in subsequent work.