24-hour evidence sprint for AI-native talent

Do not send us a polished story first. Send evidence of how you use AI to think, build, judge, and ship.

Onsite · San Jose Claude Code Codex GTM Agent Swarm

What to submit

A working artifact, sharp review, or customer-facing workflow you made in 24 hours.
A before/after artifact: what the first AI output looked like, what was weak, and what changed.
Your tool loop: Claude Code, Codex, agents, memory, evals, feedback loops, or custom scripts.
Your judgment: what you rejected, what you kept, what shipped, and what you would rebuild next.

Build a small GTM agent that researches a market, finds customers, writes a campaign, or analyzes voice-of-customer data.

Take weak AI output from a campaign, agent, landing page, or VOC summary. Review it, improve it, and explain your taste bar.

Show how you turn an AI workflow into real trust: a customer reply, operating playbook, follow-up process, or field-ready handoff.

If your submission shows A-class signal in the AI interview, you can book Hunter directly at book.paircode.ai/hunter. Interviews are in San Jose.