Purpose Driven FinTech

Banks are moving money onchain. Their data teams still can't trust what they pull from it. And every extra chain makes the bill and the mess bigger. This week I speak with Rodrigo Coelho, CEO of Edge & Node, the team behind The Graph.

We get into what trustworthy onchain data means, why data teams shouldn't have to learn crypto, how an early AI agent experiment with no spending controls ran up a $47,000 bill in 11 hours, and what happens next when tokenized assets start hopping between chains.

What you'll learn

• How Edge & Node's own infrastructure costs became its product, and the design choices behind it: plain SQL and open standards
• Why running your own blockchain nodes is not cost effective, and why buying data by the query can reach 7 figures a year
• Why data teams don't need to learn crypto to use onchain data
• Why AI agents can query onchain data in plain English, and why they still need spending controls
• Why banks waited for clear rules before going all in on digital assets
• Why tracking a tokenized asset across different chains is a new kind of data problem
• What going from employee number one to CEO, and from a big team back to a startup, teaches about building a business that makes money

Timestamps

00:00 — Banks go onchain and AI booms: what does that mean for data?

00:42 — Blockchains are just databases, and why Google Sheets explains them

10:16 — From The Graph to Amp: banks quietly running open-source code

17:21 — Where do institutions start? Pilots, client push and central bank pull

23:31 — What trustworthy onchain data really means

26:15 — The worst Rodrigo has seen: agents with no spending controls

28:27 — Why data teams don't need to learn crypto

31:34 — The real cost of blockchain data: build, buy or rent

38:25 — AI agents, SQL and tracking tokenized assets across chains

42:24 — From employee number one to CEO

44:55 — The reset: from token treasuries to a startup again

👉Connect with Rodrigo:

Edge & Node: https://www.edgeandnode.com

LinkedIn: https://www.linkedin.com/in/rodrigoco/

👉Connect with Monica:

LinkedIn: https://www.linkedin.com/in/monicamillares/

Purpose Driven FinTech. English: https://www.youtube.com/@monica_millares/videos

Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past, or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion on social media.

39% of shoppers whose card is declined at checkout don't come back. The merchant paid to get them to the checkout page, and the sale is lost, that’s a problem! This week I speak with Jack Cregan, Founder of Paymend, and we discuss recovering declined payments

Paymend gives merchants a second network once its own processors have failed a payment, opening up another 12 or 15 acquirers and routing each transaction by how the customer's issuer has performed with each one. For subscriptions, Paymend holds a failed payment for generally 30 days and bills when the card schemes signal that funds are available again.

This one is for the founder or CPO who is watching paying customers get declined at checkout and has no idea how many of those sales are recoverable.

What you'll learn

• Why 39% of declined shoppers at checkouts don't come back, and why that wastes the money already spent to win them
• Why a failed payment can still be approved by a second network of acquirers after your own stack gives up
• How routing by each issuer's history with each acquirer changes which payments get approved
• Why subscription declines are recovered by holding the payment for 30 days and billing when funds are available
• Why insufficient funds is the decline merchants can't see into, and why blind retries on paydays are the best they can do alone
• How to test a payments idea with real merchants before you build it
• Who is liable when an AI agent's payment turns out to be fraud, and why nobody knows yet

Timestamps

00:00 — Cold open: what a declined card does to a customer

02:13 — Why cards get declined in the first place

03:15 — The 39% who don't come back

08:57 — How a second network of acquirers recovers failed payments

12:34 — Subscription declines and the 30-day recovery window

16:15 — Why insufficient funds is the decline merchants can't fix alone

19:24 — Founder story: leaning on senior advisors

21:02 — Selling a company and starting again

24:59 — Test the idea with real merchants before you build

29:58 — The bet on insufficient funds, and the AI agent fraud question

31:23 — How the team uses AI, and what it costs

34:18 — The founder mindset: a network of intelligence

👉Connect with Jack:

• Paymend: https://www.linkedin.com/company/paymend-us/home/
• LinkedIn: https://www.linkedin.com/in/jack-cregan-paymend/

👉Connect with Monica:

• LinkedIn: https://www.linkedin.com/in/monicamillares/
• Purpose Driven FinTech. English: https://www.youtube.com/@monica_millares/videos
• Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past, or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion on social media.

Confused transactions cost banks billions every year. Confused customers call the bank to try to clarify something that shouldn’t have been an issue. This week I speak with Ken Hart, CEO & Founder of Snowdrop Solutions, and he makes the commercial case for a different fix: context.

What you'll learn

• Why context, not hyper personalization, is what actually fixes confused bank transactions.
• Why fixing transaction confusion pays for itself in weeks, and what it saves in call center costs.
• Why you don't have to choose between building your own AI agents and buying them off the shelf.
• Why Ken predicts a growing share of his own customers won't be banks at all within two years.
• Why starting a company to get rich is the wrong reason, and what to build for instead.

01:22 — Snowdrop's scale: 2.5 billion transactions, 200+ countries

06:36 — The $10 billion cost of confused transactions

10:01 — The one-year rule: how neobanks force incumbents to catch up

11:56 — Context, not hyper-personalization

16:37 — Build vs. buy, and a future where banks aren't the only customer

19:31 — The "wow" AI insight test: wind at your back

26:53 — The commercial model and the ROI case

43:20 — Why autonomy, not riches, is the real reason to start a company

47:59 — Bootstrapped for 12 years, and why Ken never raised VC

Connect with Ken:

🔗 Snowdrop Solutions: https://www.linkedin.com/company/snowdropsolutions/posts/

🔗 LinkedIn: https://www.linkedin.com/in/kenhart/

Connect with Monica

🔗 LinkedIn: https://www.linkedin.com/in/monicamillares/

🔗 Purpose Driven FinTech. English: https://www.youtube.com/@monica_millares/videos

🔗 Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion in social media.

Every cross border payment gets checked twice. Once by the sending bank, once by the receiving bank. Nobody questions why. This week I speak with Richard Beverley, CEO and Founder of Block Infrastructure, and he makes the case (and solution) that screening it once, not twice, is the fix the industry has been avoiding.

Block Infrastructure runs two products built on the same idea: share verified data instead of duplicating it. It streamlines KYC and KYB verification so institutions don't recheck what's already been confirmed. Its transaction screening product pre screens payments for sanctions and compliance risk using zero knowledge screening, so no bank has to expose its own customer data to do it.

We get into why AI turns ambiguous compliance "maybes" into fast, binary decisions, how a five person team plus an army of AI agents now replaces what used to take a full engineering department, why chasing niche markets beats going straight at the big banks, and the founder lesson that nobody will ever treat you as their priority, plus Richard's honest read on Founders burnout.

What you'll learn

• Why screening a payment once instead of twice cuts compliance cost and closes the FX gap that makes cross border payments expensive
• How screening lets banks catch sanctioned payments without exposing their own customer data
• How AI turns ambiguous compliance maybes into fast, binary yes or no decisions
• Why founding teams now pair a handful of senior engineers with AI agents instead of scaling headcount, cutting costs six to sevenfold
• Why not adopting AI now is the fastest way to fall permanently behind
• Why chasing niche, fast moving markets before big banks is the smarter go-to-market sequence
• Why resilience, not a great idea, is what actually gets a fintech founder through the sales cycle

Timestamps

00:29 — Why financial services is "built wrong"

04:16 — Why chasing big banks first will burn your capital

06:24 — The hidden cost: compliance gets checked twice

09:23 — $2.9 billion lost last year in compliance holds

14:24 — Why correspondent banks are disappearing

20:37 — How Block Infrastructure screens for sanctions

28:12 — Adopt AI now, or "reinvent the wheel"

35:15 — The biggest founder mistake: nobody is your priority

40:04 — Five engineers and an army of AI agents

45:51 — The honest answer on founder burnout in the AI era

👉Connect with Richard:

Block Infrastructure: https://www.linkedin.com/company/block-infrastructure/home/

LinkedIn: https://www.linkedin.com/in/richard-beverley-1044431a/

👉Connect with Monica

LinkedIn: https://www.linkedin.com/in/monicamillares/

Purpose Driven FinTech. English: https://www.youtube.com/@monica_millares/videos

Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past, or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion on social media.

60 million Brazilians can't get a credit card. For years, that meant subscription businesses simply couldn't bill them. And the moment a new payment rail shows up, everyone assumes it just steals volume from cards. This week I speak with Sebastian Fantini, Product Director of EBANX, and he makes the commercial case that a local recurring rail grows the market instead of shifting it.

EBANX connects global merchants to local payment methods across more than 20 emerging markets. It now processes transactions for 26% of Pix users in Brazil and 38% of all Pix Automático transactions. In Pix Automático's first year, 64% of the people paying with it were brand new customers, not cardholders who switched. Hotmart saw a 32% lift in customer retention after turning it on.

We get into why card subscriptions lose 20 to 30% of recurring revenue to involuntary churn and how a QR rail recovers it, why mandated adoption is what made Pix stick where other countries' schemes stalled, the go-to-market that took EBANX to 38% market share, and EBANX's plans to replicate success in SEA.

What you'll learn

• Who the 60 million credit cardless Brazilians actually are
• Why adding a local payment method can grow your addressable market instead of cannibalizing card revenue
• How mandated adoption made Pix stick where other countries' instant payment schemes stalled
• Who actually funds the 3 to 5% discount merchants give you for paying with Pix, and why they choose to
• The go-to-market that took EBANX to 38% of Pix Automático transactions
• Why card subscriptions lose 20 to 30% of recurring revenue to churn, and how a recurring QR rail claws it back
• How B2B and SaaS billing is quietly moving onto a rail built for consumers
• Why you can win a customer with marketing and still lose them at the payment checkout

Timestamps

00:00 — 60 million people, no credit card

02:26 — Who actually can't get a credit card in Brazil

04:37 — How EBANX connects global merchants to local payers

06:14 — Pix versus Pix Automático, and why recurring QR matters

09:17 — Is Pix Automático cannibalizing cards, or growing the market?

13:05 — Why Pix succeeded where other countries' schemes stalled

21:33 — Who funds the Pix discounts

23:50 — The go-to-market behind 38% market share

29:36 — Why the checkout screen decides conversion

35:01 — B2B and SaaS are paying by QR

39:24 — Reducing involuntary churn with Pix Automático

43:46 — Expanding into Southeast Asia and handling the FX

👉Connect with Sebastian:

EBANX: https://www.ebanx.com

LinkedIn: https://www.linkedin.com/in/sebastian-fantini-a0a1b970/

👉Connect with Monica

LinkedIn: https://www.linkedin.com/in/monicamillares/

Purpose Driven FinTech. English: https://www.youtube.com/@monica_millares/videos

Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past, or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion on social media.

For centuries, a voice was proof. You heard your banker, your child, your CEO, and you knew. Not anymore. A voice can now be cloned from a few seconds of public audio. The problem is not that voice disappears, it is that voice becomes easy to manufacture.

This week I speak with Anat Goldstein, Founder of FinOptima, whose whole argument is that no single signal, not a voice, not a password, not a device, can prove who is on the other end of a call anymore.

FinOptima builds explainable fraud detection for the institutions the industry skipped: community banks and credit unions, which face the same deepfake threats as the top 20 banks on a fraction of the budget and with no in-house fraud team. The numbers are not abstract. Fraud losses are projected at 40 billion dollars. In one case, criminals deepfaked senior executives on a video call and moved around 25 million, and the employee who sent it was not careless.

We get into why fraud now happens inside the live call rather than before it, what regulators mean when they demand a reason for every blocked transaction, and why the harder problem coming is not detecting fakes but proving which interactions are real.

Anat also traces the move from a 20-year banking career to founding a company, validated through customer discovery before any code, and built one hard decision at a time.

What you'll learn:

• Why a voice can no longer prove identity
• Why one signal only creates suspicion, and how voice, behavior, device and transaction history together are what create confidence
• Why fraud detection has to happen during the call, not after the money has moved
• Why community banks and credit unions carry the same threat as the largest banks with almost none of the resources
• When it makes sense to partner rather than build fraud infrastructure in-house
• What "explainable" fraud AI actually means to a regulator, and why a risk score on its own no longer holds up
• What fraud looks like once AI makes it cheap enough for anyone to run, and why trust infrastructure has to come first
• How Anat found and validated the idea through customer discovery, and what she means by resilience built one decision at a time

Timestamps

00:00 — Cold open: your voice is no longer proof

02:24 — 40 billion in losses, and a 25 million dollar deepfake call

04:13 — How do you know the caller is real?

05:21 — Why community banks and credit unions are the target

07:09 — Why today's fraud tools miss it: they work in silos

08:10 — Fraud that happens inside the live call

08:59 — What explainability means to a regulator

12:08 — One signal creates suspicion, many create confidence

17:17 — Build versus buy, and the first 90 days

20:16 — Leaving a 20-year banking career to found a company

24:37 — Resilience, built one decision at a time

27:50 — When fraud gets cheap: good AI versus bad AI

Connect with Anat:

🔗 FinOptima: https://www.linkedin.com/company/finoptima-solutions/

🔗 LinkedIn: https://www.linkedin.com/in/anat-goldstein/

Connect with Monica

🔗 LinkedIn: https://www.linkedin.com/in/monicamillares/

🔗 Purpose Driven FinTech: https://www.youtube.com/@monica_millares/videos

🔗 Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion in social media.

Banks spend millions on carbon footprint. Banks have an engagement problem. Something is not working! This week I speak with Mathias Boissonot, Founder and CEO of Handprint and he makes a commercial case for a different model: instead of tracking what your customers do wrong, quantify what they're building right.

Handprint powers ocean regeneration programs for banks like DBS, including coral reef restoration and mangrove planting that customers own, watch grow, and choose to fund with their spend. The result isn't just goodwill. It's an 875% increase in customer engagement compared to legacy carbon programs, and $23 million a year in verified ecosystem value returned to society.

We get into why Mathias measures success by a metric most founders have never heard of: the "phoenix" over the unicorn, and the sales and founder lessons he's learned building the company from scratch.

What you'll learn

• Why carbon offsetting and "your carbon footprint" features fail to change customer behavior, and what the psychology actually is
• The handprint model: quantifying positive impact instead of negative impact, and why that flips the entire customer relationship
• How banks turn nature engagement into a growth engine: referral programs with co-owned impact, nature cashback instead of standard cashback
• Why 62% of customers would switch banks over purpose misalignment, and 52% already have
• The phoenix metric: measuring company success in ecosystem value generated, not just valuation
• Mathias's biggest founder mistakes, and the best sales advice he's ever received: "implicate the pain"

Timestamps

00:00 — Cold open: has carbon offsetting failed?

04:40 — What Handprint actually does

06:55 — Phoenix vs. unicorn: a new metric for success

09:00 — $23 million in ecosystem value, every year

18:30 — Why donation and carbon-guilt features don't work

24:40 — Kill the cashback war: nature cashback instead

27:30 — From cost center to growth engine: the 875% number

32:30 — Building a company from Bali

35:30 — Every mistake we could have made, we've made

41:35 — Implicate the pain: the best sales advice he's received

44:00 — Resilience, mindset, and having fun in the pain

Connect with Mathias:

🔗 Handprint:https://handprint.tech/

🔗 LinkedIn: https://www.linkedin.com/in/boissonot/

Connect with Monica

🔗 LinkedIn: https://www.linkedin.com/in/monicamillares/

🔗 Purpose Driven FinTech. English: https://www.youtube.com/@monica_millares/videos

🔗 Purpose Driven FinTech. Spanish: https://www.youtube.com/@MonicaMillares_Español/videos

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion in social media.

What does it actually mean when an AI agent completes a purchase on your behalf? I sit down with Anatole Baboukhian, Executive Lead, Corporate Responsibility & Strategic Advocacy, and Lucy Anderson, Head of Asia Solutions, SMB Product; colleagues at Global Payments on opposite sides of the business, policy and product. We start with a plain english definition of agentic commerce and where it fundamentally changes from how we pay today. From there we talk through consent, trust, and the liability question every fintech founder needs an answer to before they build. We had a ton of fun, talk about Taylor Swift and bill payments!

What you’ll learn:

• Agentic commerce splits into two models: instructed vs. fully autonomous
• KYB still isn't solved. KYA (Know Your Agent) is next, and it's harder: identity, mandate, scope
• Liability isn't one framework: it depends where in the chain the failure happens
• Boring, high-friction use cases (utility switching, ticket buying) will win before exciting ones (travel)
• Why Marketing is about to need a second audience: agents, not just people

TIMESTAMPS

00:00 – What agentic commerce actually means

01:27 – The two use cases nobody separates

07:19 – What regulators are really asking

09:32 – KYC → KYB → KYA: the trust gap nobody's closed

16:42 – The three-layer KYA framework: identity, mandate, scope

18:44 – Why the "boring" use cases will win first

21:49 – Taylor Swift tickets vs. switching your utility provider

26:15 – Marketing to agents, not people

28:43 – Fraud, inclusion, and vulnerable consumers

32:38 – Liability: why there's no single framework

44:18 – Who will actually build these agents

49:25 – The one piece of advice for every fintech founder

👉 Connect with Lucy and Anatole:

LinkedIn: https://www.linkedin.com/in/anatole-baboukhian/

LinkedIn: https://www.linkedin.com/in/lucykanderson/

👉 Connect with Monica:

LinkedIn: https://www.linkedin.com/in/monicamillares/

Website: https://fintech-product-ai-lab.lovable.app/

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion in social media.

João Moura built a fraud detection platform operating in 188 countries not by making fraud prevention safer, but by turning it into a growth lever. This conversation gets into how organized crime runs modern fraud, why siloed data guarantees you'll lose, and how connecting institutions across borders changes everything.

What you'll learn

• Why 70% of fraud today is backed by organized crime networks, and what that means for your detection strategy (02:46)
• How mule accounts operate as layered networks across institutions and why catching one account misses the point (32:20)
• What false declines actually cost you in revenue, customer trust, and operational overhead (09:03)
• How fraud detection becomes a growth enabler when it lets acquirers onboard merchants same-day with confidence (20:19)
• Why connecting data across institutions creates a network effect that makes every client's fraud model stronger (28:40)
• What the third generation of AI-powered fraud detection looks like and why it's fundamentally different from siloed models (14:45)
• How agentic AI introduces a new authorization problem that nobody in payments has solved yet (37:15)
• What João learned about distribution, decisiveness, and sunk cost fallacies after spinning Fraudio out of a bank (46:20)

TIMESTAMPS

1:35 João's background: 20 years in AI before it was cool

2:46 How organized crime powers modern fraud

7:55 The real cost of false declines on revenue and customers

14:45 Three generations of AI fraud detection explained

16:34 How generative AI acts as a force multiplier for fraud teams

20:19 Detecting fraudulent merchants before chargebacks arrive

25:13 How much fraud Fraudio actually stops: the numbers

28:40 The Philippines rule that becomes available to 188 countries

32:20 Why mule accounts are really mule networks

37:15 Agentic AI and the unresolved authorization problem

43:30 The lazy insight that became Fraudio

46:20 What João learned: distribution, decisiveness, and sunk cost fallacies

👉 Connect with João:

LinkedIn: https://www.linkedin.com/in/mourajoao

Website: https://www.fraudio.com/

👉 Connect with Monica:

LinkedIn: https://www.linkedin.com/in/monicamillares/

Website: https://fintech-product-ai-lab.lovable.app/

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion in social media.

Everyone is talking about Agentic Payments! So I sat down with Robert Kraal, Co-Founder of Silverflow, live at Money20/20 Asia and we had a chat. Robert brings a commercial perspective to the conversation, and we focus on the what are the hard commercial questions the industry still needs to answer before agentic payments can truly scale.

We cover:

(00:00) Is the card system ready for agents?

(01:00) The commercial model nobody is talking about

(02:30) Neutral agent vs. paid agent: the Spotify analogy

(04:00) What happens when your agent talks to another agent?

(06:00) Liability: who's responsible when the agent gets it wrong?

(08:00) The tokenization gap in emerging markets

(10:00) Where to start: simple use cases first

(12:00) What Silverflow is doing to prepare for agentic transactions

(14:00) What excites Robert most about this space

(16:00) What payment professionals should be doing right now

👉 Connect with Robert:

LinkedIn: https://www.linkedin.com/in/robertkraal/

Website: https://www.silverflow.com/

👉 Connect with Monica

LinkedIn: https://www.linkedin.com/in/monicamillares/

YouTube: https://www.youtube.com/@moni_millares

TikTok: https://www.tiktok.com/@moni_millares

Website: https://www.purposedrivenfintech.com/

—

Disclaimer: This episode does not constitute professional nor financial advice and does not represent the opinion nor views of my current, past or future employers. The guest has agreed to record and release our conversation for the use of this podcast and promotion in social media.