Web3 with Sam Kamani

EPISODE DESCRIPTION 
I sat down with Chris, Chief Strategy Officer at Beldex, live at Korea Blockchain Week in Seoul. We got into why privacy isn't just a feature , it's a full-stack infrastructure problem. Chris broke down how Beldex evolved from a privacy coin in 2018 into a complete privacy layer covering messaging, browsing, VPNs, and now AI agent communication. We also got into something I feel very strongly about: context bleed in AI agents and why it's a silent killer for anyone building with multi-agent systems. If you're a founder or developer building anything that touches user data , especially in the AI or trading space , this one is for you. 
DISCLAIMER
Nothing mentioned in this podcast is investment advice and please do your own research. It would mean a lot if you can leave a review of this podcast on Apple Podcasts or Spotify and share this podcast with a friend. 
Be a guest on the podcast or contact us - https://www.web3pod.xyz/
 CONNECT 
Beldex Website: https://beldex.io/
Beldex Twitter/X: https://x.com/BeldexCoin
Beldex LinkedIn: https://www.linkedin.com/company/beldex/
Cris LinkedIn: https://www.linkedin.com/in/crisblanco/?isSelfProfile=false
Web3 with Sam Kamani: https://www.web3pod.xyz/
 KEY POINTS WITH TIMESTAMPS 
• [00:00] Chris teases the core Beldex vision: becoming invisible infrastructure that people use without knowing it
• [01:07] Introductions: Chris is CSO at Beldex, recorded live at Korea Blockchain Week in Seoul
• [01:32] What is Beldex? Privacy infrastructure that started as a privacy coin in 2018 and expanded to cover all aspects of digital life
• [02:20] How Beldex differs from Monero and Zcash , they go beyond payments into messaging, browsing, and VPNs
• [03:35] The real use case problem in crypto: too much infra, not enough apps , Beldex is building both
• [04:56] BChat: a privacy-first messaging app like WhatsApp that requires no phone number or email
• [06:52] Beldex raised an $8M round in August and is now focused on enabling builders via SDK, EVM sidechain, and a grants program
• [09:01] Message to founders: privacy should be built in from day one, not bolted on at the end
• [11:17] Technology becomes adopted when it becomes invisible , we need to stop forcing people to learn crypto jargon
• [13:13] Beldex is now testing encrypted messaging infrastructure for AI agent communication
• [14:29] Context bleed in AI agents: why privacy infrastructure is critical for anyone building multi-agent systems
• [16:37] Trading use case: context bleed between agents could be disastrous for funds and quant systems
• [19:29] Chris shares his personal tech conviction: hyper-personalised, custom-built software powered by AI is the future
• [21:29] Biggest ask: developers and founders should think deeply about why privacy matters, then come check out Beldex

EPISODE DESCRIPTION 
I sat down with Wish Wu from Pharos at Korea Blockchain Week to catch up on everything they've been building since we last spoke eight months ago. Pharos has quietly become the largest blockchain for RWA insurance and distribution in APAC, with over $116 million in real-world assets on-chain , and half of those users are coming straight from Web2. Now they're going further. Wish walks me through how Pharos is building agent-native infrastructure, solving the gas sponsorship problem for micropayments, and launching a desktop app that lets anyone , developer or not , build and run on-chain workflows with AI. With 200,000 to 300,000 agents already running on the chain, the vision is bold: more agents than humans actively transacting on Pharos within the next 12 months. 
DISCLAIMER
Nothing mentioned in this podcast is investment advice and please do your own research. It would mean a lot if you can leave a review of this podcast on Apple Podcasts or Spotify and share this podcast with a friend. 
Be a guest on the podcast or contact us - https://www.web3pod.xyz/
 CONNECT 
Pharos Website: https://www.pharos.xyz/
Pharos Twitter/X: https://x.com/pharos_network
Pharos LinkedIn: https://www.linkedin.com/company/pharosnetwork/
Wish LinkedIn: https://www.linkedin.com/in/wishlonger/?isSelfProfile=false
Web3 with Sam Kamani: https://www.web3pod.xyz/
 KEY POINTS WITH TIMESTAMPS 
• [0:00] Introduction at Korea Blockchain Week with Wish Wu from Pharos
• [0:50] Pharos has become the largest blockchain for RWA insurance and distribution in APAC with over $116M in on-chain assets
• [2:07] ~50% of Pharos users are coming from Web2 , many holding crypto for the first time
• [3:14] Pharos upgrades to agent-native support, tackling gas sponsorship friction for new users and agents
• [4:42] Why existing gas sponsorship models break down for agent micropayments and multi-chain interactions
• [6:33] Introducing programmable gas subscriptions , pay once, let your agent run freely on Pharos
• [9:22] Key challenge: closing the UX gap between Web2 apps and Web3 experiences
• [10:28] Agents level the playing field , they give Web3 a chance to compete with Web2 UX
• [11:54] Pharos is launching a desktop app , a Codex-style tool for both developers and everyday users
• [18:06] Partnership priorities: institutions and asset issuers to bring more natively issued on-chain assets
• [21:09] 12-month vision: mainnet rollout of agent-native Layer-1, more RWA distribution, and more agents than humans on-chain

I sat down with Chad from THORChain , the protocol that literally invented cross-chain liquidity , for a raw, no-fluff conversation about what it really takes to secure a decentralized exchange handling over $125 billion in volume. We dig into why security audits give people false confidence, how AI is supercharging hackers right now, and what THORChain's autonomous halt systems actually do when things go sideways. We also get into the Bitget $350M hack, what it means when funds flow through decentralized protocols, and the deeper philosophical question , where do you draw the line between censorship resistance and protecting users? Chad also shares what's next for THORChain, including Monero and Zcash support, faster swaps, and why he believes DEXs will inevitably outcompete centralized exchanges on every metric that matters. 
DISCLAIMER
Nothing mentioned in this podcast is investment advice and please do your own research. It would mean a lot if you can leave a review of this podcast on Apple Podcasts or Spotify and share this podcast with a friend. 
Be a guest on the podcast or contact us - https://www.web3pod.xyz/
 CONNECT 
THORChain Website: https://thorchain.org/
THORChain Twitter/X: https://x.com/thorchain
THORChain LinkedIn: https://www.linkedin.com/company/thorchain/
Web3 with Sam Kamani Podcast: https://www.web3pod.xyz/
 KEY POINTS WITH TIMESTAMPS 
• [00:00] Sam introduces the episode , THORChain, the Bitget $350M hack, and what's really going on with security in crypto
• [01:38] Chad joins and explains what makes THORChain different , the protocol that invented cross-chain liquidity
• [03:33] Why THORChain was the first true decentralized exchange for native Bitcoin, Litecoin, and Doge , and Monero coming next
• [04:41] Security lessons from $125B in volume , why a cross-chain DEX is orders of magnitude harder to secure than Uniswap
• [05:40] A wave of hacks hits the whole sector: Maya, THORChain's app layer, NIR, Chainflip , all within weeks
• [06:53] Why audits are the lowest standard in security , and why having an on-staff red/blue team matters far more
• [08:07] How THORChain's autonomous halt system works , the protocol can stop itself before humans even notice
• [10:36] Managing the trilemma: decentralization, security, and usability , how validator voting and pause mechanisms balance it all
• [13:32] The Terra collapse as a lesson: why you need fast-response mechanisms, not just slow governance
• [15:23] Chad's take on the Bitget $350M hack and the tension between censorship resistance and stopping stolen funds
• [16:38] Why decentralization is a requirement for liberty , and what Bitcoin gave the world for the first time in history
• [20:00] Historical precedents: the Ethereum DAO hack fork and Bitcoin's 2010 overflow bug , when protocols have rolled back transactions
• [23:13] The biggest underestimated security risk in DeFi right now: AI lowering the bar for black-hat hackers
• [25:17] Macro view: AI pulling capital away from crypto, the Genius Act, TradFi merging with DeFi, and where the cycle stands
• [27:23] What's next for THORChain , Zcash now, Monero next, more chains, faster swaps, better quotes
• [28:17] Why DEXs will inevitably beat CEXs on speed, cost, and accessibility , and why Chad is excited to be part of that future

EPISODE DESCRIPTION 
In this episode, I sit down with Denis from Humea , a product and research company that has catalyzed over $1 billion in ARR for their clients. We go deep on what it actually takes to achieve product market fit, and why most founders are looking in completely the wrong direction. Denis brings a rare dual background in physics and psychology, and he uses it to reveal how human behavior , not technology , is the real driver behind whether a product succeeds or dies. We talk about why Web3 is uniquely challenging because money arrives before product market fit, the hidden psychology of why founders resist pivoting, how Doordash generated over $300 million in additional revenue through better user understanding, what blockchain banks must do to win trust, and why simplifying your message for non-technical decision makers might be the most valuable thing you do this year. Whether you are building in DeFi, fintech, AI, or SaaS, the principles Denis shares here are universal and immediately actionable. 
DISCLAIMER
Nothing mentioned in this podcast is investment advice and please do your own research. It would mean a lot if you can leave a review of this podcast on Apple Podcasts or Spotify and share this podcast with a friend. 
Be a guest on the podcast or contact us - https://www.web3pod.xyz/
 CONNECT 
Humea Website: https://humea.com/
Humea LinkedIn: https://www.linkedin.com/company/humea
Web3 with Sam Kamani: https://www.web3pod.xyz/
 KEY POINTS WITH TIMESTAMPS 
• [00:00] Humea has catalyzed over $1 billion ARR for clients across fintech, banking, and infrastructure
• [02:49] Denis explains his dual background in physics and psychology and how it shapes his approach to product research
• [05:57] The difference between output and outcome , and why measuring the wrong thing gives you false positive data
• [08:54] Why working with an agency should always start with a friendly conversation to find genuine alignment before any contract
• [14:13] In Web3, money comes before product market fit , the complete opposite of the Web2 world
• [17:42] The IKEA effect and sunk cost fallacy explain why founders get emotionally attached and resist pivoting
• [23:08] Why mid-stage startups throw money at ads instead of fixing the underlying product and customer understanding
• [28:24] Doordash case study , course correcting the user journey generated over $300 million in additional annual revenue
• [31:45] Blockchain banking case study , turning a deeply negative NPS into a positive one through empathy-driven research
• [33:20] The four things users want from banking: returns, versatility, problem solving for their segment, and trust
• [34:04] Trust in crypto products is not about gaining confidence , it is about removing doubt
• [38:19] Why human-crafted content creates energy and connection that AI-generated content currently cannot replicate
• [43:58] Even in cutting-edge Web3, all major deals are still done human to human the old-fashioned way
• [45:24] Most B2B decision makers are not technical , your communication strategy needs to reach non-technical buyers
• [47:30] Polygon CDK dashboard case study , a brilliant engineer could not explain the interface to a colleague in another department

EPISODE DESCRIPTION 
I sat down with Rafsun, founder of Sahaj Mobile, at Money2020 to unpack one of the most exciting fintech stories I've come across in a long time. Rafsun went from managing billions in asset-backed securities at Guggenheim to building a phone financing company in Bangladesh , a market that has been starved for credit for decades. With over 58,000 customers and millions deployed entirely from their own capital, Sahaj Mobile is proving that emerging markets can outperform the West on default rates. We dig into how they used merchants, OEMs like Samsung and Xiaomi, and TikTok influencers to scale fast, how AI powers everything from fraud detection to Bengali-language customer service, and where they are headed next , including a fascinating cross-border financing model for South Asian migrant workers in Saudi Arabia and the UAE. If you are a founder or investor looking at where the real opportunities are, this conversation is one you cannot afford to miss.
DISCLAIMER
Nothing mentioned in this podcast is investment advice and please do your own research. It would mean a lot if you can leave a review of this podcast on Apple Podcasts or Spotify and share this podcast with a friend.
Be a guest on the podcast or contact us - https://www.web3pod.xyz/
 CONNECT 
Sahaj Mobile Website: https://www.sahajmobile.org/
Sahaj Mobile LinedIn: https://www.linkedin.com/company/sahajmobile/home/
Sahaj Mobile Facebook: https://www.facebook.com/SahajMobileBD
Web3 with Sam Kamani: https://www.web3pod.xyz/
 KEY POINTS WITH TIMESTAMPS 
• [00:00] Rafsun introduces Sahaj Mobile's core model , financing phones in Bangladesh with no credit score required
• [01:00] Sam introduces the episode and explains why Sahaj Mobile stands out among fintech startups
• [03:03] Rafsun's background: UC Berkeley, Kiavi (acquired by Figure), and managing $10B in ABS at Guggenheim
• [05:24] How Sahaj Mobile works: down payment, device locking, and repayment via bKash and Nagad wallets
• [06:16] Growth story: from a 100-phone pilot to 58,000+ customers since launching in 2024
• [07:09] How they got initial traction: partnering with merchants, then OEMs like Samsung, Oppo, Vivo, and Xiaomi
• [08:57] TikTok influencers and social media as a key growth channel in Bangladesh
• [12:07] How AI runs 24/7 , from loss reserve calculations to NDA reviews , augmenting Rafsun's workflow
• [13:30] AI in underwriting: ID fraud detection, face matching, and machine learning on alternative data
• [14:55] AI-powered customer service responding in Bengali using voice models and Hugging Face
• [16:49] HeyGen for AI-generated marketing videos , cutting production time to under 30 minutes
• [20:18] The SaaS and legal disruption debate , and why AI tools are actually creating more lawyer work, not less
• [27:44] Fundraising update: Series A equity round with a lead investor and a separate debt round from an emerging market credit fund
• [29:06] Expansion into MENA: financing phones in Bangladesh for migrant workers paying from Saudi Arabia, Dubai, and Malaysia
• [30:06] Next product: the Sahaj wallet with EMI discount incentives for customers who pay through it