Can We Trust AI-Written Code? Why Executable Specs Are the Future of Software ft. Gabriela Moreira

How USDT0 Moves $150M Across Blockchains Every Day

What does it take to move over $150 million in stablecoins across blockchains every single day? In Episode 90 of Colors of Web3 and Entrepreneurship, host Lam sits down with Lorenzo Romagnoli, co-founder of USDT0, the interoperability layer that has quietly become one of the most important pieces of infrastructure in the Tether ecosystem. At just 27, Lorenzo has spent his entire professional life in DeFi from LARPing on Discord during the 2020 bull run to co-founding Alloy (Tether's gold-backed stablecoin) and ultimately building USDT0 alongside the Layer Zero team. Lorenzo walks us through USDT0's meteoric rise: launching in January 2025, surviving the most stressful month of his life, and scaling from a few hundred thousand dollars a day to moving over $100 billion in total volume in just 16 months all with a team of around 20 people. He explains why Tether historically avoided DeFi, how the team finally cracked their first big partnership with Kraken's Ink chain, and why some of the world's largest hedge funds and exchanges (including an unannounced $800M Binance treasury move at 4am) now run on the rails they built. The conversation also goes deep on Tether Gold (XAUT) and why Lorenzo believes gold-backed stablecoins could one day rival the US dollar, the real product-market fit for stablecoins in Africa, Latin America, and Asia, the relentless security threats from North Korean hacking groups, his philosophy on not being greedy in crypto, and how the coming agentic AI economy could demand millions of transactions per second. Whether you're a builder, an investor, or just curious about the plumbing behind the stablecoin economy, this episode is packed with insight from someone building at the frontier. ⌛ Timestamps: 00:00:00 Episode trailer 00:00:40 Show & guest introduction 00:03:30 First steps in Web3: DAOs, Tether, and building Alloy 00:06:11 Spotting Tether's cross-chain interoperability problem 00:08:05 Building USDT0 with LayerZero 00:11:06 Lorenzo's hobbies 00:15:20 How USDT0 works 00:20:00 Some predictions and tractions of USDT0 00:25:02 The user groups of USDT0 00:30:13 Lorenzo's take on gold and Tether Gold (XAUT) 00:34:21 How Tether Gold works under the hood 00:37:05 Inside the Switzerland vault 00:42:38 Competitors of USDT0 00:45:01 USDT0's business model & the cost of security 00:50:55 The biggest challenges: hackers, pitching & iterating 00:55:21 The future of stablecoins: one winner or many? 00:59:16 Non-obvious use cases: luxury watches & private jets 01:00:52 Agentic AI, stablecoins & the millions-of-TPS problem 01:04:06 Advice for aspiring entrepreneurs 01:06:33 Where to follow Lorenzo & USDT0 01:07:30 Episode outro and next episode preview Follow Lorenzo Romagnoli & USDT0: USDT0's Website: https://usdt0.to/ [https://usdt0.to/] USDT0's Twitter/X: https://x.com/USDT0_to [https://x.com/USDT0_to] Lorenzo Romagnoli 's Twitter/X: https://x.com/zerolore [https://x.com/zerolore] Our social media: Crypto card guide: cryptocardguide.com [https://cryptocardguide.com] Twitter/X: x.com/ColorsofWeb3pod [https://x.com/ColorsofWeb3pod] LinkedIn: linkedin.com/company/colors-of-web3-entrepreneurship [https://linkedin.com/company/colors-of-web3-entrepreneurship] If you enjoy this episode, please subscribe to stay updated on our upcoming episodes with industry experts and innovators in the Web3 and Entrepreneurship spaces. Thank you for listening!

Can We Trust AI-Written Code? Why Executable Specs Are the Future of Software ft. Gabriela Moreira

In this episode, we sit down with Gabriela Moreira, CEO of Quint company and a passionate advocate for formal methods in software development. Gabriela shares her journey from compiler research and teaching formal methods at university in Brazil to building Quint — a language that makes formal specification accessible to everyday developers. She explains how Quint was born inside Informal Systems to solve a real problem: TLA+ was powerful but too hard for most engineers to read and adopt. The conversation dives into how AI is changing the economics of formal verification. Writing specs used to require learning a new language and investing significant time upfront, but with LLMs, developers can now generate Quint specifications in hours instead of weeks. At the same time, as AI writes more and more of our code, the need for confidence and correctness has never been higher — making executable specifications more relevant than ever. Gabriela also walks through real-world use cases, including AWS's use of TLA+, Circle's Arc L1 blockchain built on Quint-verified consensus, and work with Monad's BFT protocol. Whether you're building smart contracts, distributed systems, or any software where bugs can cost real money, this episode will change how you think about the gap between "it works on my machine" and "I know it's correct." ⌛ Timestamps: 00:00:00 Episode trailer 00:00:36 Show & guest introduction 00:02:04 From compilers and type systems to TLA+ and formal methods 00:06:23 Gabriela's first encounter with AI and data science 00:08:32 Gabriela’s hobbies 00:09:54 The origin story of Quint and Informal Systems 00:13:29 Deep dive into Quint and formal specs 00:19:20 Real-world use cases: Malachite, Circle's Arc, Monad BFT 00:27:42 Biggest misconceptions engineers have about formal specs 00:33:35 Bugs that traditional testing misses but Quint catches 00:38:22 Formal verification for smart contracts and Web3 security 00:41:02 Where to follow Gabriela and Quint 00:42:02 Episode outro and next episode preview Follow Quint Co & Gabriela Moreira: Gabriela's Twitter/X: https://x.com/bugarela Quint Twitter/X: https://x.com/quint_lang Quint Github: https://github.com/informalsystems/ Gabriela's LinkedIn: https://linkedin.com/in/bugarela?originalSubdomain=br Gabriela's Github: https://github.com/bugarela Our social media: Crypto card guide: cryptocardguide.com Twitter/X: x.com/ColorsofWeb3pod LinkedIn: linkedin.com/company/colors-of-web3-entrepreneurship If you enjoy this episode, please subscribe to stay updated on our upcoming episodes with industry experts and innovators in the Web3 and Entrepreneurship spaces. Thank you for listening!

Building AI for the "Third Nation" — 6M Users, $10M ARR from Singapore | Bruce Yang, Agnes AI

In this episode, we sit down with Bruce Yang, co-founder and CEO of Agnes AI, a homegrown AI company from Singapore that has already amassed over 6 million registered users and reached $10 million in annual recurring revenue all within less than a year of launching. Bruce shares how his team built a cluster of cost-efficient AI models that deliver near state-of-the-art performance at one-tenth the cost, making advanced AI accessible to the 99% who are currently locked out by paywalls and infrastructure gaps. Bruce takes us through his journey from a scholarship student arriving in Singapore at age 15, to studying math and computer science at UC Berkeley, building LinkedIn's first subscription product, launching a "Snapchat before Snapchat" social media startup in Silicon Valley, and eventually returning to Singapore to pursue a PhD at NUS right as the generative AI revolution exploded. That unique combination of startup grit, big tech experience, and academic depth fuels his bold vision for Agnes AI. The conversation goes deep into Bruce's thesis on the AI geopolitical landscape: why the world cannot afford a US-China duopoly in AI, and why Southeast Asia and the rest of the emerging world need their own sovereign AI capabilities a "third nation" in the AI race. We also explore how Agnes AI leverages post-training on open-source models, their agentic routing architecture, the future of SaaS in an AI-agent world, and Bruce's advice for aspiring entrepreneurs: keep your ego low, fail fast, and become an adaptive learner. ⌛ Timestamps: 00:00:00 Episode intro & guest introduction 00:01:06 Bruce Yang's introduction and Agnes AI overview 00:02:26 From China to Singapore to UC Berkeley 00:03:19 Working at Microsoft and LinkedIn 00:04:06 Startups experience in Sillicon Valley 00:05:03 The PhD program that eventually led to founding Agnes AI 00:07:21 The time when he built a social media startup challenging Facebook 00:09:56 Fundraising lessons and speed as essence for consumer apps 00:12:01 Hobbies of Bruce 00:15:24 Agnes AI founding story 00:17:58 Building cost-efficient models with post-training 00:21:50 Live product demo: research, slides, and character AI 00:35:54 AI inclusion and the "third nation" vision 00:42:06 Challenges facing Agnes AI: educating investors in emerging markets 00:47:17 Enterprise solutions and Model as a Service (MaaS) 00:50:42 Future of Gen AI, SaaS disruption and AGI predictions 00:56:38 Diminishing returns, GPU costs, and the road ahead 01:01:50 AI models training themselves — the evolution cycle 01:04:14 Advice for aspiring entrepreneurs 01:07:04 Where to follow Agnes AI 01:08:02 Episode outro and next episode preview Follow Agnes AI: Web app: https://agnes-ai.com/ Download on App Store: https://apps.apple.com/sg/app/agnes-ai-your-daily-copilot/id6746299015 Download on Google Play: https://play.google.com/store/apps/details?id=com.sobrr.agnes Agnes AI LinkedIn: https://www.linkedin.com/company/agnesaisapiens Agnes AI Twitter/X: https://x.com/agnesai_sapiens Our social media: Crypto card guide: cryptocardguide.com Twitter/X: x.com/ColorsofWeb3pod LinkedIn: linkedin.com/company/colors-of-web3-entrepreneurship If you enjoy this episode, please subscribe to stay updated on our upcoming episodes with industry experts and innovators in the Web3 and Entrepreneurship spaces. Thank you for listening!

The Stablecoin Guy (Ariel Eiberman): Testing 30+ Crypto Cards & The Future of On-Chain Banking

In this episode of Colors of Web3 and Entrepreneurship, Lam Trinh sits down with Ariel Eiberman, the Latin American director at GnosisPay, also known in the industry as "the Stablecoin guy". Ariel share his fascinating journey into the crypto space, which began in 2013 when he used Bitcoin as his primary banking infrastructure to facilitate international payments for polyglot meetups. The conversation dives deep into the frustrations of traditional banking, particularly in Argentina, where users face multiple exchange rates, physical bank limitations, and inefficient systems. Ariel explains why he has personally tested over 30 different crypto cards and neobanks, offering a unique perspective on which platforms offer the best fees, trust, and usability for those looking for a banking alternative. Finally, Ariel shares his vision for the future of on-chain banking. He discusses the importance of self-custody, the rise of niche neobanks, and how blockchain technology can eventually provide better financial tools for families and businesses without necessarily replacing traditional banks entirely. ⌛ Timestamps: 00:00:00 Episode trailer 00:00:34 Show & guest introduction 00:02:01 How tools and neobanks have exploded in crypto 00:02:54 The rabbit hole story, Bitcoin as a first bank 00:04:30 From ICOs to smart contract wallets and decentralized RPCs 00:06:48 Working in crypto accounting and the pain of blockchain bookkeeping 00:08:23 Why banking in Argentina is a nightmare 00:11:30 Argentina's five different dollar exchange rates 00:14:23 QR payments vs. card payments in Latin America 00:17:38 Ariel's hobbies 00:18:56 How Ariel became "The Stablecoin Guy" with 30+ crypto cards 00:21:00 How he positioned himself in a noisy Web3 payment space 00:23:58 Trust vs. fees what users actually care about 00:25:00 Why most crypto neobanks will fail in the next bear market 00:27:33 Neobanks need to go niche - The barbershop analogy 00:31:06 Best cards for zero FX fees: EtherFi, Gnosis Pay, and more 00:32:58 QR code payments with stablecoins in Bolivia 00:34:53 Vietnam's incredible banking system and crypto card opportunities 00:37:45 The future of on-chain banking 00:39:29 Some opportunities that Ariel sees for neobanks 00:42:15 Advice for people who want to break into Web3 00:43:52 Where to follow Ariel & Gnosis Pay 00:45:55 Episode outro and next episode preview Follow Ariel & Gnosis Pay: Ariel's Twitter/X: https://x.com/AriEiberman Ariel's LinkedIn: https://www.linkedin.com/in/arieleiberman/ Gnosis Pay Website: https://gnosispay.com/ Gnosis Pay Twitter/X: https://x.com/gnosispay Our social media: Crypto card guide: cryptocardguide.com [https://cryptocardguide.com] Twitter/X: x.com/ColorsofWeb3pod [https://x.com/ColorsofWeb3pod] LinkedIn: linkedin.com/company/colors-of-web3-entrepreneurship [https://linkedin.com/company/colors-of-web3-entrepreneurship] If you enjoy this episode, please subscribe to stay updated on our upcoming episodes with industry experts and innovators in the Web3 and Entrepreneurship spaces. Thank you for listening!

He Automated His Own Job and Got Into YC: Hudson Liao on the Future of Go-To-Market

What happens when a head of growth decides to automate himself out of a job before AI does it for him? You get TheHog, a YC-backed AI go-to-market platform built by Hudson Liao that listens to real buyer conversations across LinkedIn, X, and Reddit to find genuine purchase intent. Hudson joins Lam to break down his path from California firefighter to YC W25 founder, why spray-and-pray outreach is destroying brands, and what "scaling authenticity" actually means when AI is doing the work. ⌛ Timestamps: 00:00:00 Episode trailer 00:00:35 Show & guest introduction 00:04:50 Transitioning from finance to the startup world 00:07:22 The universal framework behind growth 00:08:25 Hudson's hobby 00:09:50 The origin story of TheHog.ai 00:12:07 Getting into Y Combinator and the YC experience 00:15:17 Ideal customer profile: who is The Hog built for? 00:16:50 Example of how the platform works 00:20:05 Scaling authenticity with AI 00:22:53 Traction: 160+ customers in 3 months 00:26:50 Will AI replace growth teams? 00:30:30 The future of go-to-market: agents, AEO, and omni-channel 00:31:50 Business model of The Hog 00:37:24 Advice for aspiring entrepreneurs 00:39:45 Where to follow Hudson and The Hog 00:41:11 Episode outro and next episode preview Follow Hudson & The Hog:The Hog's Website: https://thehog.ai/Hudson's Twitter/X: https://x.com/hudsonliao Our social media: Crypto card guide: cryptocardguide.com Twitter/X: x.com/ColorsofWeb3pod LinkedIn: linkedin.com/company/colors-of-web3-entrepreneurship If you enjoy this episode, please subscribe to stay updated on our upcoming episodes with industry experts and innovators in the Web3 and Entrepreneurship spaces. Thank you for listening!