Hii, I'm Saber and I'm currently 19. Sooon enough I'll be one of the biggest web3 Developers in Africa. when you Hear Saber you think of web3 development clock it.

Free research and trading AI agent on Robinhood blockchain. I'm cooking something HOT for the @openservai hackathon with @Sabercodes123. New ideas keep flooding in hours before the hackathon ends. Arena67: the ultimate research and trading AI agent on @RobinhoodCrypto is going live in hours.

A harsh reality hit me yesterday: building is the loneliest, most brutal niche in this industry. Most on CT won’t make it past the first paragraph, but for anyone actually trying to build something real, read this: I’ve traded, minted NFTs, farmed airdrops, degened & none of it comes close to the sheer mental & physical exhaustion of trying to transition from a builder into a real founder. The past two weeks have felt like a slow death sentence. Juggling 2 to 3 accelerators while fundraising will break you: > 4:00 AM pitch calls across mismatched time zones > Reworking deck slides & technical docs by 7:00 AM > Midday ecosystem syncs > Demo day presentations a few hours later. You run on 3 hrs of sleep, looping anxiety, & tired eye bags. Yesterday, we almost missed our submission for the @okx Singapore buildathon. My technical lead broke down with severe burnout & illness. I was paralyzed, ready to drop my hands lol, but the @XLayerOfficial team was kind enough to extend the portal by just a few hours. CT loves showing the funded rounds & the hackathon prizes but not the actual prize tag: 1. You lose people. You miss birthdays, cancel dinners & fail to reply for weeks. People take your silence personally & cut you off. They assume you stopped caring, without realizing you traded your entire personal life just to keep an ambitious vision alive. 2. You burn your own money, before funding ever comes, you'd burn anywhere from $500 to $1k+ every month out of your own pocket just to keep servers, nodes & infra running, before personal expenses. Over 60% of everything I've made in crypto this year has gone right back into building & with receipts to show. When you don’t own luxury cars or a real estate yet, pouring every dollar into software looks completely insane to the outside world. They might ask why is he/she not compounding? Only you have those answers. Hackathons would teach you how to ship sandboxes for quick prize money. But the ecosystems that has helped me scale up further were teams likes @XLayerOfficial, @0G_labs & @arc, they look for resilience, for founders who refuse to die. To brothers like @BunsanXBT, @dragon888 & the engineers holding it down in the TG trenches: thank you for keeping me anchored when the weight was unbearable. If you follow this journey, it’s not just for quick gains, it’s because you see the scars that come with the real upside & decided not to fold. Pressure is the admission price to reach the top 10% of industry leaders. ​Many quit. A tiny few make it out the other side. ​We keep building. I’ll see you down the road. 🌱

I'm in love @CantonNetwork @shadowdeskcc

We just submitted NimDares for the @nimiq Apps Competition Cycle 2. What if saying “I’ll do it” actually cost you something? NimDares lets you: → Create a measurable dare → Stake NIM behind it → Complete the goal → Submit proof → Get your stake back or get slashed if you quit You can go Solo, challenge friends in Team rooms, or join a public Arena. I had an argument with a friend the other day and we decided to settle it with a dare. Whoever walks over 100 steps first wins. Instead of just saying “okay, bet”, we could create the dare on NimDares, both stake some NIM, and actually have something on the line. The interesting part: the app doesn't blindly trust the UI. Your stake has to be verified on-chain before a dare becomes active, and settlement only happens after a proof verdict. We also built screenshot-based verification and a GitHub proof adapter for developer commitments. Built with Nimiq Pay + Next.js + Gemini. This one was a fun problem to engineer around: incentives, proof, escrow, and real on-chain state all in one Mini App. Try NimDares: nimdares.vercel.app Code: github.com/Saber1Y/NimDares Full demo: piped.video/1uijncjvNTQ #Nimiq #NimiqPay #BuildOnNimiq Built w/@Knowledge_JO CC: @nimiq @miniappscomp

i’m 20, and building at the intersection of ai, blockchain, and software is what i’m really obsessed with. i love understanding how things work under the hood how agents interact with wallets, how transactions move onchain, how protocols communicate, how you turn an idea into something people can actually use, and how to make the whole experience better. i work mostly with typescript, javascript, solidity, react, next.js, foundry, wagmi, and viem, while spending a lot more time lately exploring ai agents, mcp, agentic wallets, and ai-native applications. i’ve been building things around onchain payments, automated payroll, agent wallets, and other experiments where ai actually gets to interact with blockchain infrastructure. i’m also contributing to open source, participating in hackathons, learning in public, and trying to get better at understanding the systems i’m building instead of just making things work. still learning, still building, still breaking things, and honestly having a lot lotss of fun doing it.

Full-stack onchain development roadmap 2026: ➺ HTML → CSS → JS/TS ➺ React / Next ➺ Blockchain basics → Ethereum ➺ Solidity → Foundry ➺ wagmi + viem ➺ Indexer (Ponder / Envio) ➺ Account abstraction ➺ Give an agent a tool, a wallet, and a constraint ➺ Smart contracts as policy, not just business logic ➺ Ship 4–5 real systems ➺ Read the audits RepeatFrontend still matters. Contracts still matter. The new layer is agents that can act onchain without holding the spending key. Gm Builders

Been heads down building something I haven’t really talked about publicly yet. We’re building ShadowDesk a private institutional RFQ venue on @CantonNetwork . The problem: An institution wants to buy a large block of an asset without broadcasting its intent or letting competing dealers see each other’s prices. Our flow is: → Buyer creates an RFQ → Invites a set of dealers → Dealers submit blind quotes → No dealer sees another dealer’s price → Best eligible quote is selected → Asset + payment settle atomically No “trust me bro” privacy layer. The authorization and visibility rules are enforced by the Daml contracts themselves. We already have a live two-participant demo running locally, including cross-participant quote secrecy and atomic DvP. Next up: taking this from a local proof → something that can actually operate on Canton DevNet. Building this one in public. More technical details soon. @appsfactory_cc

It's still that simple bro. ➺ CryptoZombies or Cyfrin Updraft ➺ Foundry until tests feel natural ➺ Speedrun Ethereum ➺ Ship a real frontend (wagmi + viem) ➺ Ship a real indexer (Ponder / Envio) ➺ Deploy on a real L2 ➺ Read Uniswap / Aave / Morpho ➺ Read audit reports like homework ➺ Add AA + a wallet that isn't just MetaMask ➺ Give an agent a tool, a wallet, and a constraint ➺ Go to hackathons ➺ Go back to “deploy something real” and repeat

Kimi K3 What have you done to me? 😭

Give one creative brief to a media team. Make sure every role improves the next version. I built 𝐋𝐈𝐕𝐄𝐑𝐋𝐎𝐎𝐏 for the @Embody_Net Livepeer Agent Hackathon, submitting to the Livepeer Agent + OriginTrail DKG track. LiverLoop is an autonomous media production loop: Generate → Evaluate → Improve → Remember. One Director turns a creative brief into a production plan by discovering the capabilities currently available through Livepeer Agent. The Orchestrator executes the plan. The Critic evaluates the finished artifact against the brief. If something fails, the Director chooses a targeted retry instead of regenerating everything blindly. The loop can coordinate video generation, image generation, audio, FFmpeg transformations, trimming, muxing and subtitle overlays. Every capability call uses authenticated server-side access, with provider fallbacks and key rotation for reliability. The browser never receives the credentials. Livepeer returns real media artifacts, URLs and estimated costs. LiverLoop stores them, probes the files with mediainfo, checks their actual format, duration, dimensions and streams, then makes the final artifacts playable and downloadable from the dashboard. Previous versions stay preserved instead of being overwritten. The system also checks generated footage for unwanted baked-in text before allowing a version to pass. When a run completes, LiverLoop packages the brief, plan, artifacts, evaluations, retry decisions and lessons into a knowledge asset. OriginTrail DKG stores that memory so future runs can retrieve previous lessons and improve their next plan. How to use it: 1. Open the app and start a run 2. Describe your brief: format, duration, audience, style and CTA 3. Watch the loop plan, generate, evaluate and improve in real time 4. Download the final artifact, or browse earlier versions 5. Open Provenance to audit every event, cost and decision > One Director. > One Critic. > One orchestration loop. > A network of media capabilities. > Downloadable artifacts. > Explainable retries. The key word is Generate. Evaluate. Improve. Remember. Live app: liverloop.world GitHub: github.com/Saber1Y/LiverLoop full demo: piped.video/oBlLtjnV4uw Built with Livepeer Agent, OriginTrail DKG, Next.js, FFmpeg and mediainfo. #Livepeer #OriginTrail #AI #GenerativeAI

Excited to share, I joined @EqualFiLabs as a DevRel Engineer. I will be working closely with @hooftly and the team, and will be working on @StaticsProtocol . And y'all not ready for what we are cooking at Statics Protocol. What is Statics Protocol? Statics Protocol is financial infrastructure for existing on-chain assets for everyone who lives on-chain. Mainnet Phase 1 announcement incoming, stay tuned. Join the TG community for updates - t.me/EqualFi/1

Update on this; just incase you might need help — it's sorted, and credit where it's due. The TL;DR nobody gives you: Render/Vercel couldn't work for my app at all. My hackathon build runs a self-hosted OriginTrail DKG node that binds to localhost, and my app's Director loop runs in the same process so it can reach that private socket. PaaS apps can't see another machine's loopback expose the node publicly and I'd be leaking a private knowledge node just for a free URL. Dead end. So to the person who said "just use Zeabur" @TuoyoS26091. You were onto something. In the end Zeabur is what got me across the line, pairing with a Tencent VPS (~$3/mo, fits my $5) that accepted the payment DigitalOcean kept failing with "card declined" (Grey virtual card). Node + app on one box under systemd, node stays on 127.0.0.1, real on-chain publishes. Lesson for anyone else building an agent/node thing: when your app is private infrastructure and not just a website, "deploy on Render" is a dead end, but the right host + a cardfriendly VPS gets you there. It's live now clean HTTPS URL before I submit. l ink: zeabur.com — enter my referral code "ayodejiakintobi1" at checkout to enjoy 10% off!

I need help 🙏 For a hackathon I need a VPS (DigitalOcean / similar) to host an OriginTrail DKG node for an always-on demo flow. My Grey virtual card keeps getting rejected by DigitalOcean - "card declined" with no reason. Same check probably fails everywhere. Which VPS providers accept payments from Nigeria? Which ones work with Grey cards? Anything ~$5/month is fine. Open to: alternate card services, prepaid-friendly hosts, crypto payments, or any workaround that isn't jumping through DO support hell. #web3 #vps

15+ Open-Source Repositories at the Intersection of AI Agents & Web3! The most interesting part of agentic software isn't the AI, it's giving agents autonomous permission layers, wallets, and onchain identity. Here is the complete builder directory for AI × Crypto: Agent Wallets & Payment Protocols: > Coinbase AgentKit: github.com/coinbase/agentkit… > SDK (Great Onchain Agent Toolkit): github.com/goat-sdk/goatSpen… > Agent Spending Control: github.com/Saber1Y/SpendaSky… > AI Payment Protocol: github.com/skyfire-xyz > Onchain Agent Frameworks & Infrastructure Eliza OS Framework: github.com/elizaOS/elizaAuto…

Selctive Privacy @CantonNetwork

Gm Builders. Lets WIN this weeekk

I desperately need to get on a podcast. 😭 Not even necessarily because I have some grand story to tell. I just genuinely love having deep, intellectually stimulating conversations. I get bored so easily with conversations that never go beyond surface-level stuff. Give me someone who can challenge my ideas. Someone who can disagree with me without making it personal. Someone who can say “I don’t think you’re right, and here’s why” and actually make me reconsider my position. I love conversations where we’re both trying to figure something out rather than trying to “win” an argument. Talk to me about technology, society, philosophy, human behaviour, business, creativity, religion, the future, why people do the things they do, ideas you completely disagree with, or some random thought you’ve been obsessing over for three weeks. I especially appreciate people who are emotionally intelligent enough to separate “you challenged my idea” from “you attacked me.” Honestly, intellectual conversations are one of my favourite ways to escape boredom. I want to leave a conversation knowing something I didn’t know before, questioning something I thought I knew, or seeing something from a completely different perspective. So yeah… Someone please put me on a podcast. I have thoughts. You have thoughts. Let’s argue respectfully for two hours and accidentally learn something. 😭

Sunday Space. Topic: Can we trust AI agents with money? We’ll talk: > Smart wallets / ERC-4337 > ERC-8004 agent identity > Onchain policy instead of “the model decides” > What breaks when agents get memory If you’re a dev, founder, or just trying to understand AI x crypto, pull up let’s have a chat. Time: 8pm WAT No slides. Just talk + questions. nitter.net/i/spaces/1aKbdEq…

Jsyk, it was not a FREE entry hack. i can remember seeing the hack, tbh but the UI took all my attention tbh forgetting it was literally an hackathon.