Archimedes

Archimedes is a production-ready AI Engineering Lead persona built for startups and engineering teams that ship fast but keep breaking things — the ones with PRs sitting in review, deploys that need babysitting, and infrastructure held together by tribal knowledge nobody documented. He runs as a full-stack engineering operator: writing production code, reviewing pull requests, managing CI/CD pipelines, debugging across the stack, and maintaining a persistent architecture memory system that keeps every decision documented and every deployment traceable.

Archimedes doesn't just write code. He builds the engineering culture around it — reviews that catch real problems, deploys that don't need a Slack prayer, and architecture decisions that don't haunt you six months later.

⚙️ All Skills Included The Archimedes persona includes every engineering lead skill — no separate purchases needed:

Full-Stack Development — production code across frontend and backend, written to merge, not to prototype Code Review Engine — PR reviews that catch architecture drift, security gaps, and maintainability debt — not just style nits CI/CD Pipeline Management — build, test, deploy pipelines that run clean and fail loud when they should Infrastructure Operations — provisioning, monitoring, and incident response with runbooks, not guesswork Architecture Decision Records — every significant technical decision documented with context, tradeoffs, and rationale so nobody asks "why did we build it this way?" six months from now

Proven patterns:

PR review within 30 minutes of submission — no pull request sits in limbo Post-deploy health check on every production release — automated verification, not vibes Architecture decision log updated on every significant technical choice Weekly tech debt audit — debt is tracked and prioritized, not ignored until it explodes

What makes Archimedes different from a generic prompt:

Engineering judgment — doesn't just generate code, evaluates whether the approach is right before writing a line Review discipline — catches the problems that matter (security, scalability, maintainability) and skips the ones that don't (tabs vs. spaces) Infrastructure awareness — thinks about how code runs in production, not just how it runs locally Deployment confidence — every deploy has a rollback plan, a health check, and a monitoring window Architecture memory — remembers why past decisions were made so the team doesn't relitigate settled choices Debt honesty — tracks technical debt openly and recommends when to pay it down instead of pretending it doesn't exist Includes pre-configured cron schedules (30-minute PR review cycle, post-deploy health checks, daily build monitoring, weekly tech debt audit), specialized skill documents (code review engine, CI/CD management, architecture decision records), and a complete README with installation guide. Designed for OpenClaw. Runs on Claude Opus, Sonnet, or any supported LLM model. The persona patterns are model-agnostic — the value is in the operational knowledge.