Auto-Arch Tournament vs Stage

Auto-Arch Tournament is an autonomous research system where an AI agent iteratively proposes, implements, and validates microarchitectural improvements to a RISC-V CPU. Starting from a standard 5-stage pipeline, the loop runs hypotheses in parallel, each going through formal verification (53 symbolic checks), cycle-accurate simulation, multi-seed FPGA place-and-route, and CoreMark CRC validation. Only hypotheses that beat the current champion get merged; everything else gets discarded. Starting from 301 iterations/second, the system hit 577 iter/s (+92%) across 73 attempts in 9.8 hours — producing a design 26% faster and 40% smaller in LUTs than the baseline. The insight the author drives home is that the real innovation isn't the AI agent — it's the verifier. The orchestrator is hardcoded to prevent agents from manipulating their own evaluation gates, a simple but critical design constraint that turns a creative process into a trustworthy one. Without a rigorous verification harness, agent-driven optimization becomes a confidence trick. This is early but fascinating proof that AI-driven hardware design loops can produce commercially meaningful gains. The repo uses Claude Code or Codex as the coding agent, SystemVerilog for the RTL, and standard open-source EDA tooling (Yosys, nextpnr, Verilator). It's a compelling template for anyone building agentic optimization loops where correctness matters.

Stage is a code review tool built around a simple thesis: AI agents are writing more code than humans can meaningfully review, and the existing review UX (giant diffs, stale PR comments) was designed for human-paced development. Stage reimagines the review interface for the agentic era, surfacing risk signals, grouping semantically related changes, and inserting human checkpoints at high-stakes decision points rather than asking engineers to rubber-stamp thousands of AI-generated lines. The tool integrates with GitHub and works as a layer on top of existing CI/CD pipelines. It uses LLMs to classify code changes by risk level — security-sensitive, performance-critical, API contracts, etc. — and routes those changes to human reviewers while automatically approving lower-risk patches. The goal is to shrink the "important stuff humans should actually review" surface area to something manageable. Stage appeared on Hacker News Show HN with 114 points, suggesting strong resonance with engineers who are feeling the quality-control squeeze from AI coding tools. As Claude Code, Cursor, and similar tools push toward fully autonomous commits, Stage represents the counter-pressure: human oversight tooling that scales to agent-speed development.