Andy Fan

Needless to say, @KaitoAI is the worst launchpads recently: > Terrible price performance > Projects delay TGE for quarters, with no clarity on how Kaito or the teams are using users’ funds during that time > Vesting schedules and terms change quickly without clear reasons ICO season is far away. Scam season is always here.

I think JobHopin explored this idea before - using AI to scan resumes, match them with current JDs, and suggest improvements. But they shutdown their business 2 years ago from my opinion, to grow the DAU, you could go one step further: add a live recommendation feed of JDs (from X or any sources u have). That way, users don’t just optimize their resume, they also discover real opportunities in just a the same place.

I vibe coded a private resume builder to help job seekers Goal was to build: >> Easy, keep-your-data resume builder (your data stays yours) >> Real recruiter feedback as if it was coming from me Match your resume to a job you're applying to with: >> Fit score explanation - not just "70% match" but WHY and what to fix >> Gap analysis with action items - "You're missing ABC, here's how to bridge that gap" Project sitting in the graveyard. Should I actually ship this, or is this a solved problem?

Who's responsible for the current Defi chaos

@0xAbhiP Most apps do exactly the same thing They hope us gonna forget to cancel and at least get the 1st month subcript

why does every saas product have a 14 day free trial that requires a credit card upfront? just give me the trial and let the product convince me

TVL ❌ Total value locked ✅ Total value left Wtf did i just see here?

Defi is cooked! The Defi world had time to stabilize after the @DriftProtocol exploit - and now it’s taken another hit. This time, the issue comes from @KelpDAO , where a serious security flaw in its token bridging mechanism (built on @LayerZero_Core) led to a massive unintended mint of assets due to compromised key governance. Here’s the breakdown - and why this matters more than it looks on the surface. > They didn’t hack the code or exploit a smart contract > They poisoned the nodes that verify transactions, while those same nodes reported everything was normal > With nothing flagging the attack, $290M was taken undetected KelpDAO had one verifier when they needed multiple. LayerZero says they warned them. That move turned the problem into systemic risk - leaving lending protocols like @aave exposed to bad debt while the attacker exits efficiently. Right now, the situation is still evolving. Most major Defi protocols have insurance mechanisms to absorb losses, and mitigation efforts are already underway But zoom out - this isn’t just a bug. It’s a pattern. Recent attacks are no longer targeting smart contract logic - they’re targeting governance layers. We’ve spent years hardening contracts, formal verification, audits… and attackers simply moved one layer up. Because governance is softer. Messier. Human. Weak key management. Over-trusting third parties. Shipping fast over. Securing properly. That’s the real vulnerability. Right now, humans are still the weakest link in the system. And incidents like LayerZero x KelpDAO don’t just cost money - they ripple across the entire DeFi stack.

@0xvietnguyen You miss $MMT

Crime of Crime Projects have manipulated prices: MYX Finance (Sep 2025) ChainOpera AI (Oct 2025) DeAgent AI (Nov 2025) PipPin (Dec 2025) River (Jan 2026) PipPin (Feb 2026) Siren (Mar 2026) RaveDao (Today-Apr 2026) Don't let yourself become a victim

@ProofofIntern Could u grant me the access?

swipe file updated (v2)! if you're a beta user of the swipe file v2, you can now see this week's updated content. this includes: - april fools content - social campaign ideas - easter content - memes - event ideas - campaign/promo ideas want access to the beta? comment below and i'll send over a code 👇

@HakResearch @buidlpad Họ có details rewards từng ngày nên double check APR dễ bác ạ. Khổ market lỏ đi farm mà còn bị tính reward sai. Em ban đầu cũng tính farm bên Buidlpad đó cơ mà thôi thấy APY flexible thì qua đây fix APR còn hơn, trộm vía dự án đang trả đúng

@AndyJfann_ @buidlpad thanks thông tin từ bác!

CẢNH BÁO: BUIDLPAD ĐANG TÍNH SAI TIỀN CHO AE! Farming trên Concrete & Bybit cũng có lỏ nhé ae! Nền tảng @buidlpad đang tính toán sai lợi nhuận cho người dùng và bản thân mình mặc dù đã đóng góp cho team trên kênh Discord nhưng đội ngũ MOD có vẻ không hiểu vấn đề gì đang xảy ra. Cụ thể: 1⃣ Một khoản gửi 12.000 $USDT vào từ ngày 31/03/2026 2⃣ Tính đến ngày 03/04/2027 thì đã nhận được tổng cộng $5.9 lợi nhuận dựa trên 3 ngày. -> Mỗi ngày nhận được $1.96 lợi nhuận. 3⃣ Tuy nhiên nếu tính trên APY khoảng 12% (chuyển đổi sang APR 11.39%) thì LỢI NHUẬN MỖI NGÀY TỐI THIỂU LÀ $3.74. Nghĩa là sau 3 ngày lợi nhuận tối thiểu phải là $11.22. - Lợi nhuận Buidlpad đang trả: $5.9. - Lợi nhuận thực tế phải nhận được tối thiểu: $11.22 Lưu ý: 12% thấp hơn con số ban đầu là 16% và thấp hơn cả con số hiện tại trên Buidlpad là 12.42%. Liệu có phải lỗi hiển thị? Thực ra đây không phải lỗi hiển thị đến từ Buidlpad bởi vì mình đã thử rút tiền ra khỏi nền tảng thì con số mình nhận về dự kiến cũng chính là con số hiển thị. Thời gian gần đây thì mình cũng gặp phải một số các nền tảng Yield Farming hiển thị APY một kiểu nhưng khi farm ra thực tế thì con số nhận về thấp hơn rất nhiều. Gần nhất chính là @ConcreteXYZ khi mà hiển thị APY đâu đó từ 7 - 8% nhưng thực tế nhận về chỉ rơi vào khoảng 3 - 4%. Nên ae cầm $USDC $USDT để tối ưu thì nhớ tính toán, quan sát cho cẩn thận tránh bị nền tảng móc túi trong một thời gian dài. Một vụ nữa là @Bybit_Official khi có mức phí rút tiền lên tới 0.1% cho $USDT trên Mantle Vaults của họ. Mặc dù mình đã hỏi đội ngũ Bybit trước đó thì nhận được phản hồi là phí rất thấp, không đáng kể tuy nhiên khi mình rút tiền thì bị tính phí cao nhất là 0.1%. Đi Farming đã gặp nhiều rủi ro về Hack nhưng bên cạnh đó cũng có cả ty tỷ vấn đề phát sinh.

Retention in Web3 is a critical problem every founder needs to take seriously. It’s not just about good UX or incentive campaigns. It’s about designing a sustainable business model where users can earn and stay engaged over the long term, not just during short-lived hype cycles. When users feel like they’re part of something bigger, they stick around, and they bring others with them. That’s when you know you have something scalable.

DeFi is dead? CT has been throwing this take around lately. Honestly, it’s not totally wrong. Yields are compressing below 5%. Defi risks are anywhere: so many hacks & depegs happened just in Q1 2026. And liquidity is fragmented across too many chains and protocols, making farming like a sprint competition between different platforms The old playbook doesn’t work like it used to. Back in early 2020 era, DeFi sold a very clear dream for literally with a decent fund to farm: stable income, 2-3-digit APY with looping, and potential airdrops Today? You’re taking more risk for less return. The next phase of DeFi won’t be driven by new farming platform, ponzi loops, or “innovative” yield mechanics It will be driven by better underlying assets. We’re already seeing the shift like T-bills, private credit, tokenized equities, and more which will be looking for capital efficient ways to become more productive assets. The protocols that win won’t be the ones offering the highest yield. They’ll be the ones that attract and optimize high-quality capital. If you’re still farming like it’s 2021, you’re probably choosing the wrong path.

@shindrom_ We shall so many cases just like that

@AndyJfann_ Great breakdown. Especially agree that liquidity isn’t just a growth lever but core infrastructure. Too many teams still optimize for short-term TVL spikes without thinking about retention or real usage, which eventually backfires.

@0xdocterZ That's why i call it dead capital. Money in without triggering the business model will erode anw

@AndyJfann_ Strong points here. One thing I’d add is that net-new liquidity should be a key metric, not just raw TVL. Recycling capital through incentives can look good short-term but doesn’t build real depth. Retention + quality of liquidity matters more than headline numbers.

@rayansadri You could do the same stuff with some free tool and chatgpt rather than pay 99$/month for this shit

Tried it out. So the product scans stuff, then it shows Reddit tells me people post about my company, surfaces competitors I already knew about ages ago, wraps it in a dashboard, and wants me to scream “OMG we saved $60k. At this point I’m convinced half of startup hype is just polished demos. How’s this the “ultimate CMO” for me

Your AI CMO? RIP your marketing team and agencies? All it does is summarize websites, check SEO, and give a few suggestions, and one recommendation tweet about hiring an AI CMO? $99/month? Thanks, but no thanks.