PIR schemes tailored to the Ethereum hot state can and archival history, allowing users to read chain data from remote servers without revealing what is being queried. The sharded approach allows optimizing for different data types, contexts of usage, and tolerance to latency.
New scheme. Sub-second preprocessing on a 32 GB database and ~30 ms response time. Productionizing with GPU acceleration via external collaborators.
Flesh out the multi-engine sharded architecture with ≥2 inaugural schemes; builds on our sharded PIR design for Ethereum state.
End-to-end demo of ETH balance retrieval over LeanPIR, targeted at wallet integration.
Reproducible harness comparing PIR schemes via a mix of reported data, independent replication, and benchmarking under unified test vectors.
Port VIA from Pythonic specs to a Rust reference implementation; feeds into the sharded harness and benchmark comparisons.
Academic paper on LeanPIR and its GPU acceleration. EuroTT workshop keynote targeted for May 2026.
Benchmark the leading candidates for GPU-accelerated PIR — Inspire, Onion v2, LeanPIR, VIA, and Harmony — at 100+ GB scale. Benchmark matrix being db/entry sizes, db segmentation, no. GPUs, ..
End-to-end sharded PIR with ≥2 schemes over a non-trivial amount of Ethereum state data. Stretch carried from Q2.
Technical post on snarkifying archival state to reduce database size, enabling smaller PIR servers.
Exploring PIR served from Ethereum nodes and how that will work depending on the statelessness route taken by the protocol.
| KPI | Target | Status |
|---|---|---|
| % of Ethereum state servable via PIR | Under practical overheads | Ongoing, various promising schemes for different slices |