Backend Developer

Position: Backend Infrastructure Engineer — Matching Engine & EVM Node

Stack: Rust · RocksDB · EVM / Ethereum JSON-RPC · Kafka · Redis · Postgres · WebSocket / Centrifugo · AWS / Kubernetes · libp2p

The Role

We're hiring a backend infrastructure engineer to work on two core systems: Cura Blockchain, where you'll lead its adaptation to EVM RPC, and our Rust Match Engine, where you'll extend it from spot to perpetual futures (Perps). This is high-performance, stateful systems work where correctness, determinism, and recoverability matter as much as speed.

You don't need to have done both before. Solid CS fundamentals, fast learning, and strong ownership are required — Perps and EVM specifics can be learned on the job.

What You'll Do

Build the Ethereum-compatible JSON-RPC layer for Cura Blockchain (eth_*, blocks / transactions / receipts / logs, eth_call, subscriptions) and make it work with real tooling: MetaMask, ethers / viem, Foundry / Hardhat, explorers, and indexers

Map Cura's native state and RocksDB storage to EVM concepts (accounts, nonce, storage, gas, finality), and own node / RPC performance and load testing

Extend the match engine to Perps: positions, margin / leverage, funding, mark / index price, liquidation, and ADL — while preserving deterministic replay, snapshotting, and crash recovery

Add perps state, risk checks, and settlement integration across Kafka, Redis, and the real-time data path, without hurting hot-path latency

Operate what you build on AWS / Kubernetes, and profile before optimizing

What We're Looking For

Solid CS fundamentals (hard requirement): data structures & algorithms, OS, networking, databases / storage, computer architecture — you reason from first principles about complexity, concurrency, memory / cache, and consistency

Fast learner: you can read a spec and a codebase, build a mental model of an unfamiliar domain quickly, and ship correct work

Strong ownership: you own design, implementation, deployment, and production issues end-to-end, and fix root causes

5+ years in backend / infrastructure with Rust, C++, or Go in performance-sensitive systems (trading, blockchain nodes, databases, or similar)

Distributed systems experience: ordering, idempotency, snapshotting, and replay-based recovery; production Kafka / Redis experience

Strong Plus

Deep understanding of Reth and/or go-ethereum (Geth) internals — execution, state trie, txpool, sync, RPC implementation — e.g. you've read the source, run/modified a client, or contributed

Deep understanding of Solana — Agave / Firedancer, Sealevel, accounts model, PoH, consensus (Tower BFT / Alpenglow), SVM, and RPC performance

Prior work on an EVM compatibility layer, perps / futures engine, risk or liquidation system

Bonus — AI-Assisted Development

We use AI coding tools daily. A considered point of view here is a plus: how you structure context, specs, and verification so agent-generated code (Claude Code, Codex, etc.) is safe to merge into a critical system, and where you think AI helps vs. silently hurts. A short paragraph in your application on what you've tried will stand out.