Product Design, UI/UX Design, Mobile App Design, Front-end Development
Timeline
Feb 2023 - Dec 2025
Role
Lead Designer & Front-end Developer
Bondex job portal Desktop and Mobile
When I joined Bondex in early 2023, we had just crossed 1 million users. Today, we're at over 5 million. I want to share what that ride looked like from the design seat.
What is Bondex?
For those who don't know, Bondex is a Web3 professional network. Think LinkedIn, but built for the crypto and blockchain world. Users can find jobs, build their reputation, and actually earn rewards for referring people who get hired. The platform now has over 2 million profiles and works with companies like Binance, Chainlink, Solana, and CoinMarketCap.
I came in as Senior Product Designer in February 2023. At the time, the team was building out the core experience, getting the mobile app stable, and preparing for what would become a major year of growth. By late 2023, we launched the Job Portal, which completely changed what Bondex could do. Instead of just being a network, we became a place where real hiring happens.
I designed and helped code the job portal interface from scratch. The goal was simple: make it easy for Web3 companies to post jobs and for users to find and apply to them.
We also built the referral system that lets anyone earn bounties for successful hires, along with the Recruiter dashboard where recruiters can manage their referrals, track how candidates are progressing, and add notes. Some of those bounties go up to $20,000.
Getting Promoted
In October 2024, I became Lead Designer. By then we had grown from around 1.5 million to 5 million users. My role shifted from just designing features to owning the whole product experience across web and mobile.
I also started doing more front-end development. Bondex has a small team, so wearing multiple hats is normal. I worked on Angular for the web platform and Ionic for the mobile apps, which let us ship to both iOS and Android from one codebase. I used Cursor to speed up my coding, but my years of coding experience meant I could always read and review what was being written. When something needed to ship, I could help build it, not just design it.
Bondex App main pages
How I Work
I design with delivery in mind. When I'm working on a feature, I'm already thinking beyond the mockup. What should the copywriting say? What are we actually trying to achieve with this feature? What are the goals and success metrics? Can we A/B test different versions? Once the MVP is live, how many users are engaging with it and what is the data telling us? These questions shape how I approach the work from the start.
I also tend to explore multiple concepts rather than jumping straight to a final design. A Figma file might have three or four directions for the same feature. I share these early with engineers and product to get feedback before committing to one path. It surfaces problems sooner and gets the team aligned before anyone's written a line of code.
I don't hand off a pretty mockup and hope for the best. I hand off something the engineering team can actually build without a hundred follow-up questions. And most times, I could jump into the codebase myself to help ship it.
Solving Real Problems: AI Applicant Scoring
One feature I'm particularly proud of is our AI Applicant Scoring system. Let me walk through how we approached it.
The problem: Companies posting jobs on Bondex were getting overwhelmed. A single role could attract hundreds of applicants, and many weren't even qualified for basic requirements. During user research, one participant from Nigeria shared that over 90% of his direct applications never got a response. He assumed he didn't make it, but couldn't even tell if his application was ever seen. The system was broken for everyone. Companies drowning in noise, candidates disappearing into a void.
The solution: We built a two-layer screening system. First, geographic filtering. If a job is NYC-only and someone applies from a non-matching location, they're automatically filtered out. Second, AI scoring that evaluates every applicant against the job requirements on a 1-5 star scale. Only 3-5 star candidates move to the Screening step for hiring manager review. 1-2 star applications are auto-rejected with a response, so candidates at least know where they stand.
Why it matters: Hiring managers now see only relevant candidates instead of sifting through hundreds of mismatches. Candidates get actual responses instead of silence. It's not just efficiency. It's respect for everyone's time.
Recruiters confirmed this from their side too. One participant told us she was actively evaluating new tools because her current ATS lacked AI screening. She'd dealt with the volume problem firsthand and was ready to switch platforms for a solution. That validation helped us know we were solving a real pain point, not an imagined one.
AI Review tab on a job application details page
What We Didn't Solve: The Reputation Gap
Not everything went as planned. Our Reputation system is a good example of building something valuable but realizing the real problem was harder than we thought.
We built verification for wallet activity, LinkedIn tenure, and other credentials. Users could prove they'd been on LinkedIn for years or show their crypto wallet history. In user research, participants told us they were more likely to engage with verified profiles, especially in Web3 where bots are everywhere.
But the same research revealed what we were missing. Users told us directly: "A key verification would be work history and skills verification." They wanted to prove they actually worked somewhere. And more importantly, prove their skill level. I can say I know React, but what's my level? Junior? Senior? Expert?
We shipped workplace verification as a first step. Users can now verify they worked at a company through email domain matching, with manual review for edge cases.
One problem we did solve was identity fraud. During research, a recruiter told us she'd had three fake video interviews. Real calls, but the person was using some kind of face filter that was "semi-obvious," and their LinkedIn photo didn't match. In a fully remote, global, pseudonymous industry, this is a real risk. Our v1 Reputation system addressed this through KYC identity verification. When a candidate is verified, companies know the person applying is who they claim to be.
But skill-level proof remains unsolved. It's the harder problem, and we knew it. Sometimes you ship what you can while acknowledging what's still missing.
Reputation page on the Bondex App
Designing with Research
I ran user research sessions throughout my time at Bondex. Sometimes with Figma prototypes, sometimes live testing features in production. In May 2023, we ran moderated sessions with recruiters to validate the Job Portal before launch. In 2024 and 2025, we interviewed users from Slovenia, Belgium, Ukraine, Romania, and Nigeria to understand how they were actually using the platform.
These sessions shaped real product decisions. The insight about candidates never hearing back directly informed our AI Applicant Scoring system, which now auto-responds to every application so candidates know where they stand.
Other sessions revealed that users didn't want wallet activity visible on their profile because it made them a target for scams. We'd assumed it was a feature. They saw it as a risk. We added show/hide controls in response. But the same research showed nuance: one user said he wouldn't display token transactions, but would share something like a course completion credential. That helped us think about verification as selective, not all-or-nothing.
We also heard repeated complaints about stale job listings we hadn't noticed internally. Users from multiple countries flagged the same issue. And a user from Ukraine told us he didn't understand what the Reputation score actually did for him—a reminder that features we think are clear often aren't.
Research isn't something I do once and forget. It's how I make sure we're solving real problems, not just the ones we imagine.
Not just a designer
I mentioned wearing multiple hats, so here are some numbers. Over my time at Bondex, I made over 1,600 GitHub contributions across the web platform and mobile app repos. That includes commits, pull requests, and code reviews. I wasn't just pushing pixels in Figma and handing things off. I was in the codebase, shipping features alongside the engineering team.
My Github contributions through the years
I also got hands-on with testing. I reviewed builds on both web and mobile, catching issues early before they reached QA or users.
Our QA lead Devanka once told me that "this kind of quality ownership from a designer is rare". For me, it just made sense. If I'm going to design something, I want to know it actually works the way I intended.
Features I Shipped
Over almost three years, I designed and helped build a lot of what users see today. Some highlights:
•The Job Portal, including job listings, application flows, and the referral bounty system that lets users earn up to $20,000 for successful hires
•The Recruiter dashboard where recruiters can manage referrals, track candidates through the hiring process, and add notes
•Company tools including post a job, view applications, archived jobs, talent pool search, company profile setup, and team member management
•Talent features like view applications, referrals dashboard, and verify employment experience
•Bondex internal tools to manage job applications, referrals, workplace verification, flagged talent, talent pool, and company profiles
•AI job matching that recommends roles based on your resume
•Daily Streaks with bonus wheels, badges, and rewards to keep users coming back
•Public profiles so users can see each other's credentials and reputation
•The Bondex ID and Reputation system, which puts professional identity on-chain with verified credentials from LinkedIn, GitHub, and crypto wallets
•Job alerts and the "Jobs for you" personalized recommendations
•The 2025 Bondex TGE page
•Most recently, the talent onboarding flow for users coming over from Web3.Career after the acquisition (not live)
The Big Milestones
2024 was huge. We raised $10 million in funding led by Animoca Brands and Morningstar Ventures. We hit over 1.5 million monthly active users. We onboarded major clients. And we kept shipping features.
In 2025, things moved even faster. We launched the $BDXN token on Binance Alpha. We acquired Web3.Career, the number one Web3 job board on Google, which added 1.7 million monthly visitors and 100,000+ talent profiles to our ecosystem. I led the design for the integration, which you can read about in the Web3.Career case study. We hit 300,000 on-chain attestations for verified credentials.
What I Learned
Working at a startup that grows 5x teaches you to ship fast, even when things aren't perfect. As a startup still finding product-market fit, you're constantly building new things. You don't always get the chance to circle back and polish something until much later. If at all. That's a trade-off you have to accept.
I also learned how much users understand things differently than you expect. Research sessions consistently surprised us. You can't assume what makes sense to you makes sense to them.
And I learned the value of being a designer who can code. When I could jump into Angular and help implement what I designed, it saved time and made the final product better. When I could spot issues in a build before it reached QA, it kept the whole team moving faster. There's no lost-in-translation moment when you understand the full product delivery process.
What's Next
After almost 3 years at Bondex, I'm ready for my next challenge. I'm looking for a role as a Product Designer, Design Engineer, or something that lets me do both design and development. I'm open to any industry where I can bring this hybrid skill set.