Oct 16, 2026 · 5 min · 4 segments
Most teams building custom banking software start in the wrong place — the feature list. This episode of DEV.co cuts straight to the three foundational decisions that determine whether a financial…
Start with the core, because this is the decision that you genuinely cannot walk back.
The core banking platform is your ledger of record.
It holds accounts, posts, transactions, calculates interest, and emits the events every other system in your stack reacts to.
Your mobile app, your admin console, your fraud pipeline, your data warehouse.
All of it is downstream of this one choice.
Change the core later, and you are essentially rewriting the bank.
Three cloud-native vendors dominate the shortlist for new builds right now.
Mamboo is the composable SaaS option, favored by neobanks and fintech lenders that want to move fast and configure products rather than write code from scratch.
ThoughtMachine's VaultCore sits lower in the stack.
It uses what they call smart contracts to define each financial product and code, and it has Tier 1 deployments at institutions like Lloyd's and Standard Chartered.
Then there's FinxAct, now inside Pfizer, which tends to be the pragmatic answer for U.S. regional and commercial banks that need a cloud-native real-time core with proximity to U.S. charter requirements.
The real split is this.
Mamboo if you want something live in months and you're comfortable working within the platform's abstractions.
Thought Machine if you need ledger-level control and have the engineering depth to maintain smart contracts in production.
And you should know that implementation costs trend significantly higher there because of that depth.
Thinks act if you're a US bank and you want the pragmatic path.
And build your own ledger only if you have a genuine reason the vendors cannot satisfy.
A novel asset class, extreme throughput, something structurally unique.
A bespoke double entry ledger in Postgres with event sourcing is a two year project before it's boring enough to trust.
and boring is the goal.
Even once you've picked a managed core, you'll write substantial code.
The services that typically live outside the core include an onboarding orchestrator, a KYC decisioning layer, a fraud and transaction monitoring pipeline, a notification service, statements and reporting, and an open banking facade.
The pattern that works is a thin API gateway in front of domain services communicating over an event bus that the core publishes to.
The anti-pattern to avoid is treating the core like a database and writing ledger-shaped logic in your own services.
Start with the core, because this is the decision that you genuinely cannot walk back.
The core banking platform is your ledger of record.
It holds accounts, posts, transactions, calculates interest, and emits the events every other system in your stack reacts to.
Your mobile app, your admin console, your fraud pipeline, your data warehouse.
All of it is downstream of this one choice.
Change the core later, and you are essentially rewriting the bank.
Three cloud-native vendors dominate the shortlist for new builds right now.
Mamboo is the composable SaaS option, favored by neobanks and fintech lenders that want to move fast and configure products rather than write code from scratch.
ThoughtMachine's VaultCore sits lower in the stack.
It uses what they call smart contracts to define each financial product and code, and it has Tier 1 deployments at institutions like Lloyd's and Standard Chartered.
Then there's FinxAct, now inside Pfizer, which tends to be the pragmatic answer for U.S. regional and commercial banks that need a cloud-native real-time core with proximity to U.S. charter requirements.
The real split is this.
Mamboo if you want something live in months and you're comfortable working within the platform's abstractions.
Thought Machine if you need ledger-level control and have the engineering depth to maintain smart contracts in production.
And you should know that implementation costs trend significantly higher there because of that depth.
Thinks act if you're a US bank and you want the pragmatic path.
And build your own ledger only if you have a genuine reason the vendors cannot satisfy.
A novel asset class, extreme throughput, something structurally unique.
A bespoke double entry ledger in Postgres with event sourcing is a two year project before it's boring enough to trust.
and boring is the goal.
Even once you've picked a managed core, you'll write substantial code.
The services that typically live outside the core include an onboarding orchestrator, a KYC decisioning layer, a fraud and transaction monitoring pipeline, a notification service, statements and reporting, and an open banking facade.
The pattern that works is a thin API gateway in front of domain services communicating over an event bus that the core publishes to.
The anti-pattern to avoid is treating the core like a database and writing ledger-shaped logic in your own services.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.