Go back
The HR Tech Platform Rebuild: Scaling Calyptus from <10K to 120K Users
Leading the Product Design Turnaround that Drove 12X+ User Growth and Realigned Market Fit for a Recruitment Platform.
Rebuilding an MVP from Scratch to Win 80K+ Candidates
Before summer 2023, Calyptus was barely standing - a buggy MVP, under 10K users, and no real product-market fit. After a strategic pivot, the founders brought me in to rebuild the product from scratch and deliver a second MVP in just six months.
Our goal: attract 80K qualified candidates post-launch to spark growth on the recruiter side.
This case study focuses on the candidate experience.
Owning the Entire UX: Candidate + Recruiter Sides
I led end-to-end product design, from research to dev handoff. I worked closely with:
3 co-founders
A product manager (focused on engineering)
Back-end and front-end developers
I was the only designer and the go-to person for decisions, alignment, and quality across both sides of the product.
Research That Exposed the Real Problem: Candidates Felt Ignored
User Interviews
We started with five interviews with users of the v1 platform. Despite the small sample, the message was clear and painful:
Desk Research
To go deeper, I layered in desk research:
Case studies of successful recruiting tools
Candidate psychology research
Real job-seeker stories from LinkedIn
A key blind spot emerged: the post-application journey. Candidates submitted applications and then… disappeared into a void. To fix this, I ran an internal workshop that led to a new chat feature — complete with recruiter templates — to make feedback faster and more transparent.
Trust, confidence, and energy were the core barriers. Our redesign had to restore all three.
4 Pains That Shaped the Product Strategy
As a result of the research, I defined primary user pains I should focus on:
Being Ignored: Candidates were ghosted after applying.
Slow Hiring: The process dragged on for weeks with no clarity.
Constant Proof: They had to repeatedly prove their skills and history.
Early Mismatches: It was hard to know early on if a company was a good fit, wasting time.
Redesigning the Candidate Experience for Clarity, Speed, and Match Accuracy
I started with a lo-fi drafts based on research and business goals.
Then, I conducted review sessions with the team, evaluating designs against 3 key criteria:
Are main actions obvious?
Can devs build this quickly?
Does this serve a business goal?
Based on the feedback:
Redesigned job cards for instant clarity before applying
Cleaner layouts to help users scan quickly
Key requirements moved to the top to reduce time wasted on mismatches
Job preferences hidden to surface only essential info
Matched skills made prominent to boost confidence and transparency
Clearer CTA wording so users always knew their next step
Design System
To speed up development, I built a modular design system inspired by Tailwind UI - fully documented, fully reusable.
Result?
5x faster idea-to-handoff time
Dev team shipped 2x faster post-launch
This system became the backbone of our velocity.
Fixing the Drop-Off, the Low Matches, and the Trust Gap
Onboarding Drop-Off (60% loss → ~25%)
The onboarding process had a 60% drop-off rate; only 31% of users completed profiles.
I cut onboarding to 4 essential steps, surfaced it after showing value (job matches), and placed it in context rather than upfront.
Profile completion jumped from 31% → 49%.
Drop-off shrank from 60% → ~25%.
Low Job Matching Rates
Poor matches discouraged users and hurt retention.
2 changes drove improvement:
Higher profile completion (above) boosted algorithm quality.
Removing the “low match” badge eliminated the demotivating effect, while still showing match percentages.
Intimidating Verification Flow (28% → 65%)
Experience verification felt rigid and formal, especially for current roles, with only 28% completing verification.
I added self-verification as an option - less pressure, less friction, more progress.
Verification jumped from 28% → 65%.
Match accuracy improved significantly.
Result: 120K Users and New Recruiting Partners After Launch
We exceeded the 80K acquisition goal in 6 months, reaching 90K, then 120K shortly after. The new version helped the team land new recruiting partners and begin scaling the business.
The Constraints That Shaped Every Design Decision
Little research time. Devs were already coding when I joined.
Two user personas at once. I designed for both candidates and recruiters. Few candidate insights, so I pulled from external research; more recruiter context came from our internal hiring team.
A tiny cross-functional team. Balancing alignment vs. speed was a daily challenge.
Being the only designer. Every decision required data, not opinions — I had to advocate constantly and clearly.
Test Earlier. Follow the Money Sooner.
Build space for testing, even when it feels impossible.
Guerrilla testing helped, but not enough. Small testing windows early on would have prevented some late surprises.
Candidates mattered, but recruiters were the revenue engine.
Our research showed recruiters ultimately funded the product. In the next iteration, I shifted more attention to recruiter flows to align design with business sustainability.
How I Transformed Calyptus's Most Complex Screens:
Get in touch
I’m always interested in exploring new opportunities, collaborating, or exchanging ideas with like-minded individuals. Feel free to email me or connect on LinkedIn if you'd like to discuss an open product designer or ux/ui designer position or a potential project.