Sunny: an AI job-search companion I designed, built, and shipped solo.
A job-search companion I designed and coded solo: an AI job scout, interview prep, and Gmail automation, built to reward effort instead of outcomes. It's live, and I use it every day.

Impact
• Gmail automation files "thanks for applying" noise and surfaces the emails that turn into interviews. • The AI scout finds fresh matching roles and skips jobs you've already applied to.
Scope
A live, installable web app built solo in about a week of evenings (Next.js, Supabase, and a $0-budget AI architecture).
Timeline
2026
Role
Designer, PM, and AI-assisted engineer: solo, end to end (product strategy, UX, visual design, front-end, deployment)
The problem: job-search tools measure what you can't control
Most job-search trackers count interviews and offers, the two things a candidate can't control. I wanted a tool that kept momentum up instead.

Process: AI as my engineering team
I treated AI as my engineering team: ship a feature in Claude Code, use it in my own search that day, and iterate from real friction. Every product decision stayed mine.

The insight: reward effort, never outcomes
Reward effort, never outcomes. You control applying, not replies, and the most anxious moments, like rejections, deserve the warmest design.

The design: one companion that grows with you
One companion character that grows with your effort: XP, streaks, and a badge you can only earn by getting rejected and continuing.

The result: a live product I use every day
A live, installable app: AI-picked jobs, a color-coded pipeline with bulk editing, interview prep built from your own story, and a Gmail sweep that files the noise.

Takeaways
AI-assisted development changes what one designer can ship. The craft shifts from pushing pixels to making sharp product calls and testing them live.