← BLOG

2026-01-09

THE REFINEMENT FLYWHEEL: HOW I SHIP WITH AGENTS

tl;dr: Iterate on plans, not code. Front-load the arguing, let multiple models poke holes, only implement once there's nothing left to debate.

After a year of pairing with agents daily, the biggest shift wasn't prompt engineering—it was workflow structure. The gap between folks who get consistent results and those who bounce off isn't model quality. It's knowing when to stop prompting and start planning.

→ KEY TAKEAWAYS

→ THE WORKFLOW

Step 1: Brainstorm

Start with a high-level explanation of what you're trying to achieve. Work back-and-forth with the agent to design the solution.

Step 2: Design Review

Run 1-2x to iterate on the plan.

Step 3: Make Tasks

Convert the plan into atomic work units.

Step 4: Improve Tasks

Run 1-2x until quality.

Step 5: Execute

Go for a walk.

Step 6: Code Review & More Tasks

See what the swarm comes back with.

Step 7: QA

Put on your QA hat. Actually use the feature.

Step 8: Final Review

After all that, review the code once in GitHub.

→ A COUPLE NOTES

Tasks are for the agents, not for you. We still use Linear to manage bugs, features, work. Tasks just let us break work into small, unambiguous pieces that an individual agent can pick up.

Watch the context window. For Opus, I keep individual sessions under 70-90k tokens. Above ~130k, you start getting dumb decisions. If I'm getting over that, I stop and ask for a summary. Then open a new session:

We were working on X—here's a summary. Use extended thinking and familiarize yourself. Let's continue...

Feedback loops are critical. Give your agent a way to interact with your software. A linter, a type checker, a test runner. Results will be 2-5x better—especially in implementation. Give them the ability to know when they're making a mistake and self-correct.

If an agent makes a bad choice, call it out. Tell it to write in claude/agents.md to not make this mistake again. Doesn't always stick, but it does help.

→ THE TOOLS

The principles above are tool-agnostic. But if you're curious:

→ THE REPO

I keep my agent skills and workflow configuration in a public repo: github.com/Vpr99/agent-setup

→ THIS SEEMS LIKE A LOT OF STEPS...

→ WHAT I'M ALSO THINKING ABOUT

Making a "real" executor (aka a bash loop) and maybe replacing Beads with a simple prd.json file. Read: Tips for AI Coding with Ralph Wiggum