full-stack blockchain developer | Serial builder| founder @pexaapp | developer @poratlops| building the next unicorn
Joined July 2021
- Tweets7.7K
- Following1.8K
- Followers2.6K
- Likes16.1K
Pinned Tweet
Join The wait-list
The Pexa waitlist moves.
Every friend you bring pushes you up the line.
And we're quietly working on something for the people who bring the most.
Not saying what. Just don't be last.
pexaapp.xyz
Chris retweeted
Smart contracts are brilliant at doing things.
They’re terrible at telling anyone about it.
So we built PortalOps for the Portaldot Hacker House ⚡
An event layer that helps applications reliably understand what happens on-chain.
Point it at a Solidity or Rust contract 👇
Chris retweeted
Here is my submission for the #Gamdom100kRound2 Challenge
A Casino Bar where a Gamdom Boss plays Gamdom cards and introduces the platform to the Bar Owner
Building the concept of what @gamdom is all about… grab your popcorns 🍿
@gamdom @stevewilldoit @kingadukes
OfferCheck helps job seekers make safer, more informed career decisions in a world where fake job offers and recruitment scams are increasingly difficult to recognize.
Users can paste a job offer, recruiter message, or employment opportunity into OfferCheck to receive a structured investigation report highlighting suspicious payment requests, pressure tactics, missing employer information, unusual compensation claims, and other potential warning signs. Each report explains the evidence, provides practical next steps, and helps users understand what they should verify before accepting an offer or sharing sensitive information.
Designed around clarity, accessibility, and user privacy, OfferCheck aims to make recruitment safety available to everyone without requiring an account just to investigate an offer.
Our long-term vision is to evolve OfferCheck into a broader recruitment trust platform, combining AI-assisted analysis with real company and domain verification, job-listing checks, recruiter reputation signals, community-reported scam intelligence, and fraud-pattern detection. Over time, OfferCheck could help job seekers, employers, recruitment platforms, and career communities build a more transparent and trustworthy employment ecosystem.
OfferCheck does not simply label opportunities as safe or fraudulent. It helps people understand the evidence, identify what remains unverified, and make better-informed decisions before they take the next step.
offercheck-job-offer-jyqb.bo…
Stop taking advantage of people @gainzalgo
Hello @gainzalgo I was announced and DM-ed as the 3rd place winner for your design contest and I forwarded my BEP20 address in DMs as requested more than a week ago.
Haven't received a reply or payout yet. Would appreciate a quick look into this. Thanks!
Noted 📝
It's actually nice and I'm checking it out I tried out the derivatives
I was asked about the power rule
Where if y=x^n ,dy/dx=nx^n-1
I wrote that but it said I was wrong and gave my that dy/dx x^n =nxn-1
So my conclusion the model is great, and if possible should be able to read answers and relate it with its own answer find similarities with probability of them been similar.
I built RecallMap for the Arc microgrant hackathon. It's a learning app where you don't just study, you can prove you learned it.
Here's what it does and how it's built.
THE PROBLEM
Most studying is rereading and highlighting. It feels productive and you forget it in days. And when you finish, nobody can verify you actually know the thing. Certificates are just claims.
WHAT IT DOES
→ Upload a PDF, notes or a doc (up to 25 MB). The server extracts the text.
→ AI drafts flashcards and quiz questions from YOUR text. Every item has to cite a passage and quote evidence from it, then a second blind pass fact-checks it against the source. Nothing is saved until you approve it.
→ Answer-first study: you type your answer before the card flips. Retrieval is what builds memory. Then you rate it and spaced repetition (SM-2 style) schedules the next review.
→ Quizzes are graded on the server, so answer keys never reach the browser.
→ A concept map shows what's strong, what's due and what's fading.
HONEST BADGES
Recall Ready needs two separate quiz attempts at 80%+, at least 30 minutes apart. Topic Master needs 90%+. No cramming, no participation trophies. XP lives in an idempotent ledger with daily caps, and every rule sits in the database, not in the UI.
ARC INTEGRATION
When you earn a mastery badge, you can publish it as a credential on Arc mainnet.
→ I wrote and deployed a Solidity contract, CredentialRegistry, with an owner and issuer roles.
→ The credential data is canonicalised (sorted keys, random nonce) and hashed with keccak256. Only the hash goes on-chain, never your name, notes or answers.
→ Issuing is a job state machine: issue, then confirm. A credential is only marked "verified" after the transaction receipt succeeds and the on-chain hash matches ours.
→ Every credential gets a public /verify page that re-reads the contract live and runs 4 checks: data matches fingerprint, recorded on Arc, hash matches on-chain, issued by the authorised issuer and not revoked.
→ Gas is paid in USDC by the issuer wallet. About $0.0014 per credential. Arc's sub-second finality made it feel instant. The issuer wallet's balance is guarded and issuance has a daily cap.
THE STACK
Next.js 16 and React 19, Neon Postgres with Drizzle, Auth.js, Cloudflare R2 with signed uploads, OpenAI for generation, Resend for email, viem for chain calls, and a durable jobs table with a cron sweep for anything that needs a retry. 89 tests, including the contract running on a real in-process EVM.
Try it: recallmap.top
Contract: 0xd14aca8cd7331d4d5d2b91a7a3afd0edef0b7369 (Arc mainnet)
A real credential transaction: explorer.arc.io/tx/0x6ff7995…
Make learning something you can prove.