Payments Platform at Scale

The Challenge

The Payments Platform was mid-migration from a monolith to microservices architecture, keeping the system that processed billions in transactions annually running without taking it down. Reliability and security posture needed work. Stripe had set a goal of supporting 100 payment methods globally, but the platform interfaces needed to enable that expansion were still being built, leaving global teams bottlenecked. Planning and execution across the 400+ engineer org needed better structure to move at a faster pace, and with two new leaders coming in simultaneously, alignment needed to be rebuilt. The TPM function was new to Stripe, cross-functional partners were still learning what to expect from it, and the team I inherited was still building its technical depth.

Approach

• —Built the TPM team from 5 to 26 in one year, with leaders embedded across each major platform area: Payins, Money Movement, Payouts, Reconciliation, and Billing. Partnered with the VP of TPM to define level expectations and built the interview framework, educating engineering leaders on TPM hiring loops so they were partners in the process. Used the existing team as educators for incoming hires, giving them a stage while solving the onboarding bottleneck.
• —Transformed planning and execution for the Payments Platform by establishing a cross-organizational collaboration framework that aligned US and global teams on quarterly roadmaps, resolved technical dependency conflicts, and reduced continuous interruption to engineering work. The model was adopted by teams beyond Payments Platform.
• —My team drove the 100 payment methods program and global payment method enablement at the platform level. We defined clear ownership between platform and global teams, built documentation frameworks for self-help that freed up engineering resources, educated leadership on the end-to-end steps from regulatory permits to coding to shipping, analyzed architecture pain points, and initiated the move toward shared platform services and no-code options.
• —Led reliability and security posture hardening across the platform. On reliability: rate limiting was being treated as ops work, not product work. Built the case with payments engineering leaders, worked with the CTO on prioritization, formed a tiger team, and drove execution. A mindset shift that took a full year to land. On security: migrated services to a container framework, implemented least privilege access controls, closed merchant data access for expired and closed accounts, and built a UI tracking framework for visibility into progress across teams.

Results

• —TPM team built from 5 to 26 in one year, with leaders embedded across all five major platform areas.
• —Planning and execution model transformed for Payments Platform and adopted by teams beyond it. Cross-organizational framework aligned US and global teams on quarterly roadmaps, reducing dependency conflicts and engineering interruptions.
• —100 payment methods program unblocked through ownership definition, documentation, architecture streamlining, and shared platform service investment.
• —Reliability and security posture strengthened across the platform. Rate limiting elevated from ops work to product-level priority across 50+ teams with CTO alignment, alongside container migration, least privilege access, and merchant data access controls.

What I Learned

• —Define what good looks like before you hire. Scaling a team 5x in a year is possible, but only if the bar is set before the first offer goes out. Consistency at speed requires upfront investment in level expectations and interview design.
• —The function you build shapes everything that follows. Establishing TPM credibility at Stripe meant showing what a technical, embedded TPM could do that hadn't been seen before. The first few hires and first few program wins set the tone for the entire function.
• —Culture change is the hardest program to run. Convincing 50+ teams to treat reliability as product work, not ops work, required a year of patience, CTO-level visibility, and finding the right engineering leader to own it alongside you.
• —Remote-first requires more intentional structure. COVID made everything harder. Building cohesion, onboarding fast, aligning across time zones — none of it happens by default when you can't be in the same room.

What Others Say