Senior Backend Engineer  ·  Distributed Systems  ·  Blockchain Infrastructure  ·  Developer Platforms

~16 years shipping production systems. repeatedly the founding / early engineer who sets technical direction, builds the standards and infrastructure other engineers build on, and takes products from zero to one.

EVM-compatible ZK-rollup L2 with AI-powered, sequencer-level security. i was one of the first few engineers, building the protocol alongside the team from inception to mainnet. my work centered on the org's engineering foundations -- co-architecting the open-source CI/CD across Zircuit's repos, establishing the release-management and incident-response processes the team ran on, and introducing CodeRabbit as the org's AI code-review standard. the protocol is built on Optimism Bedrock.

public code from that work:

• zkr-monorepo-public -- Zircuit's modified op-node, chain-ops, and Bedrock contracts (Go / Solidity) -- the rollup's consensus and chain-operations layer
• l2-geth-public -- the L2 execution client (modified geth)
• zkr-go-common-public -- shared Go libraries underpinning the Zircuit services

took Macro from early engineering to launch -- set up the backend and devops practices, built public SDKs and API surfaces, and stood up the prototyping infra that shipped multiple products inside a year.

• launch on Product Hunt

• xerrors-ts (ts-xerrorof) -- tagged domain errors for TypeScript. define a factory per domain, narrow with isErrorOf, wrap causes, remap across layers, and use the same shape whether you return a Result or throw

• Mattermost / playbooks -- authored the OpenAPI spec for the workflows REST API, which became the documentation template other teams followed  
• Mattermost / playbooks -- designed and built the official workflows Go client (SDK), with an API modeled on the Stripe and GitHub client conventions to stay intuitive and hard to misuse  
• verifiable contribution history -- my Mattermost PRs  ·  my playbooks PRs

reliability at the edges of distributed and agentic systems -- typed error boundaries across service layers, structured-output schemas with validation and re-prompting loops, and eval sets that catch regressions before they ship. building backend agentic systems with the Claude API, MCP, and orchestration patterns.

Languages & Frameworks

Data & Infrastructure

AI & Agentic

Web3

open to senior / staff backend and infrastructure roles -- remote, contract or full-time.

• Email
• LinkedIn