krishn

I am Full stack dev who can wear multiple hats, previously i was core team member of beezie.io which is marketplace of collectibles (e.g. pokemon, basketball, one piece cards and many more),

What beezie does?
- Those cards widely popular in US / Japan markets, We tokenize physical card/asset and turn it into digital form (aka NFT but tradable) with help of Brinks, so it's marketplace where user can buy/sell/bid their digital cards.

I built, optimized and owned end-to-end implementation of the Claw Machine for Beezie.io which generates $100k in 2 months (thanks to marketing team). The Claw Machine is an interactive Web3 gamification tool where users pay to "play" and receive random cards with varying swap values. Users have a 3-minute window to either swap the card for money (Crypto) or keep it. It was designed for Web3 enthusiasts and "degen" traders to increase platform engagement and revenue. 

I collaborated with design team and blockchain team to meet an outcome, Developing interactive UI (it's first impression where user attracted to spend money and play claw), Connecting the frontend to Hono.js/Elysia.js APIs and handling Flow (its blockchain where all assets lies, calculating randomness of claw machine cards, making onchain transactions) interactions via OpenSeaPort (provides SDK of buy/sell mechanism of NFT) and Managing deployments across Vercel, Cloudflare, and eventually migrating codebases to Railway.

When building the foundation for Beezie.io, one of the most pivotal product decisions I made was choosing Privy.io for our authentication layer over more traditional options like Wagmi or RainbowKit. While standard wallet connectors are the industry norm, they present a massive barrier for "Web2-style" users who aren't familiar with seed phrases or browser extensions. By choosing Privy, I was able to offer social logins (like email and Google) that automatically generated "embedded" wallets for users. This significantly lowered the barrier to entry for the Claw Machine. The trade-off was a higher integration complexity and a subscription cost, but it was a necessary investment to ensure our user growth wasn't throttled by the friction of blockchain onboarding.

As the platform scaled, I had to make a high-stakes technical pivot regarding our infrastructure. We were originally hosted on Vercel, but we began encountering a critical "middleware bypass" issue that threatened our security during high-traffic periods, particularly our "PROD days" when revenue peaked. I had to choose between fighting Vercel’s specific architectural limitations or migrating the entire codebase to Railway. I ultimately chose to migrate to Railway because it offered more granular control over the server environment and eliminated the edge-case security risks we were facing. Although we traded off some of Vercel’s polished frontend deployment features, this move provided the production stability required to protect our $100k+ bi-monthly revenue and gave us peace of mind during peak trading hours.

Our development cycle was often shaped by high-stakes external pressure and volatility. During critical phases, internet outages caused by protests in Turkey disconnected our lead developer, Berke, forcing me to step up and take sole ownership of both frontend and backend production pushes to keep the platform live. This was especially grueling because our "Key Days" fell on Saturdays and Sundays; these windows generated the majority of our revenue, making any bug or downtime financially devastating. Operating under this "zero-room-for-error" constraint required me to maintain extreme technical discipline while shipping updates in a high-pressure, solitary environment.

Knowing what I know now, I would have implemented Mixpanel from Day 1. We wasted time trying to fix Google Analytics' inconsistencies between environments and miscalculations of sales.