Testing an AI Code Review Workflow
sterlingcrispin · x · 2026-07-10
The author mentions using Sol Ultra in /goal mode, waking up to massive code changes from +90k to -5k lines, and then requesting an HTML exec update.
Their conclusions were:
- The visual design was terrible (design slop), but
- As a code review tool, this method is highly valuable.
- Because the iteration speed is so fast, the agent generates massive changes before you can fully understand them, and you converge on the final version through review afterwards.
Overall, the post shares an agentic coding workflow experience centered around "rapid generation + human review."
Related event: AI Agent Refactors 90,000 Lines Overnight(2 posts)→
More from coding & agent
- Soft Clamp cuts tool-call overuse in multi-teacher distillation, from 13.7% to 9.0% — antgroup · 2026-07-21
- Agent harness memory loss and compaction are still a major usability problem — adityaag · 2026-07-21
- SpecJudge runs locally on Ollama to pick the right-sized AI model for your project — jokiruiz · 2026-07-21
- A developer maps out six design rules for CLIs that humans and AI agents can both use — yujiezha · 2026-07-21
- A coding-agent skill that forces ADHD-friendly, answer-first output — ayghri · 2026-07-21
- A set of agent skills for CAD, robotics, and hardware design — earthtojake · 2026-07-21