The role:

You’ll be a senior engineer on the UniFi stack, working across both its offchain infrastructure and onchain components.

You do not need prior experience

with this exact stack. We hire for distributed-systems depth and proven ability to ramp fast on unfamiliar deep tech. If you’ve spent years in Ethereum L2s or L1 clients, wonderful, but the bar is ramp speed, not resume keywords. If your background is trading systems, consensus, or low-latency backend infrastructure and you can learn the Ethereum-specific parts quickly, we want to talk.

What you’ll build and own:

• The Engine (Rust): L2 and L1 preconfs (preconfirmations), parallel block building, and blocksapce configuration.
• Cross-chain settlement paths: vault contracts custodying capital on Ethereum L1, execution on the L2, and relayer services with confirmed-send semantics, serialized nonce management, and crash-restart recovery.
• On-chain infrastructure: Rollup Contracts and messaging between L1 and L2.
• Off-chain services in TypeScript (Node 20, viem): high-cadence pricers signing EIP-712 payloads, cross-chain relayers, risk monitors, keepers, and quoting bots, engineered for restarts, nonce races, and stale-data edges.
• TEE attestation infrastructure: hardware attestation verification (Intel TDX DCAP, AMD SEV-SNP) implemented in both Solidity and Rust, kept in lockstep by cross-language parity tests.
• ZK Proofs: Zero knowledge proofs
• The test surface that keeps it honest: Foundry fuzz and invariant suites, adversarial regression tests, and live-testnet E2E harnesses.

You’ll pick a primary layer; you’re expected to read across all of them.

What we’re looking for (must-haves):

• Excellent general software engineering: 5+ years (or equivalent evidence) building production systems other people depended on. You write code reviewers trust and tests that catch real bugs.
• Strong distributed-systems fundamentals: You reason natively about ordering, idempotency, at-least-once vs. exactly-once, partial failure, retries, and recovery. You can walk us through a system you built where one of these was the hard part we will ask, in detail.
• Proven fast ramp on unfamiliar deep tech: Point to a time you went from zero to shipping in a domain with a real learning curve: a consensus client, a database engine, a compiler, a codec. The domain matters less than the slope.