May 2, 2024
A shared operating system for HR, recruiters, and candidates
A recruiting platform that turns HR, recruiters, and candidates into one shared workflow — designed 0→1, shipped, and in use across Taiwan, Vietnam, Singapore, and Japan.
B2B SaaS
0→1 → Funded
APAC · 4 Markets
HR Tech
Role
Sr. Product Designer
team
2 Product Designers, 3 Engineers ( FE, BE, AI )
platform
Web
Timeline
1 year (ongoing iteration)
BACKGROUND
LeapIn is a cross-border recruiting firm — the platform began as a fix for its own business. Hiring across borders still runs on email and scattered files: no one knows which job description is current, updates die in mail threads, and the same questions come back every week. That fog kept the firm's own team from growing. The goal: earn a place in a crowded ATS market by putting client HR and recruiters on the same platform, so alignment happens inside it — shorter hiring cycles, not one more tool to maintain.
MY ROLE
Led product design from MVP to scale, defining scope directly with founders (no PM)
Drove product decisions through research with in-house and external recruiters, and client HR
Designed the shared pipeline model for HR, recruiters, and candidates
Defined the brand guideline and design system, keeping all visual output consistent
RESULT
Launched in Q3 2024 and in active use by partner companies and recruiting firms across Taiwan, Vietnam, Singapore, and Japan. By the team's estimate, hiring for a single position on LeapIn took 51% less time and 29% less cost. This traction supported the company's fundraising.
Time isn't spent finding people — it's spent aligning information.
Research kept showing the same pattern: the hours went into reconciling versions of the same information — not into sourcing or interviewing.
The leverage is alignment, not ATS.
On the opportunity map, pipeline tools were already crowded; a single source of truth for job information had high impact and few competitors.
Transparency is political.
Companies needed pipeline visibility to keep sending cases; recruiters couldn't show detail without losing leverage.
HR, recruiters, and candidates share one product — different workflows, different permissions, opposite needs for transparency. How does each role see only what it should, without turning transparency into an on/off switch?
Design carried the roadmap. Founders had features they wanted first, and every investor demo brought back a new ask.
SOLUTION
Re-architecting the recruiting flow into an AI-powered shared ecosystem, integrating 4 key Agent tools into a seamless human-AI collaborative loop.
Scope as strategy
The MVP focused on becoming the single source of truth for JD information — deliberately deferring talent pool browsing, recruiter recommendation, and market analysis to later phases. Information consistency offered the highest impact and differentiation, and was the data foundation the deferred features would depend on; building them first would have left them sitting on top of fragmented data.
Information architecture as permission model
The data model used a three-tier hierarchy — Company → Job → Candidate
One status, two views
The biggest pipeline decision was designing recruiter and HR as parallel views over a shared status engine.
A candidate has one underlying status, but each role sees a role-specific label and a different next action — the recruiter sees "Waiting for HR feedback"; HR sees "Review CV" — at the same moment, for the same candidate.
This resolved the transparency tension surfaced in research: both parties feel they own the workflow, while what they see is in fact a controlled view. The same logic runs down to field level — on the same candidate profile, the recruiter keeps contact details, CV, and edit rights, while HR sees professional information and the decision entry point.
It also made cross-pipeline task handoff workable — uploading a CV created an HR task, returning feedback created a recruiter task — keeping the feedback loop inside the system rather than across email.
