I owned checkout and built recurring billing across Salesforce and three processors. Periodic reconciliation caught missed or out-of-order webhook events; I also brought Salesforce ownership in house.
Responsibility and context
Employment: Technical Lead & Salesforce Platform Owner · November 2022–December 2024. I joined LegalBrave, a consumer legal-tech startup founded by Protégitas, as Development and Integrations Lead. Two external consultancies, one for Salesforce and one for web development, reported to me for delivery. After LegalBrave’s wind-down, I moved to Protégitas with the same charter.
Protégitas provides legal-services memberships to companies. I took ownership of Salesforce requirements and prioritization, alongside web application development and cloud integrations. This was platform ownership before my formal Technical Product Manager role at Marsh.
Building the recurring-billing capability
I owned the website checkout end to end and integrated Salesforce with PayPal, Stripe and Adyen. In our implementation, Adyen tokenized cards and charged on demand; the subscription lifecycle needed its own billing logic.
I built the recurring-billing engine: subscription state, billing cycles and charge scheduling. AWS Lambda webhook handlers streamed payment events into Salesforce, including successful and failed payments and new subscriptions.
Webhooks alone were not sufficient to keep the CRM aligned with the processor. A periodic reconciliation job re-checked processor state against Salesforce to catch missed or out-of-order events. Chargebacks detected in the event stream were recorded in Salesforce and triggered an alert to the call-center team responsible for working the dispute.
Delivered capability: integrated checkout, recurring billing, webhook event handling and periodic processor-to-CRM reconciliation. Conversion and recovery impact is unquantified.
Choosing the delivery path
I shipped integrations first with Zapier and Pipedream for speed, then rebuilt them as production services on AWS. The decision was about getting a useful integration released and moving it to services I could operate as the implementation developed.
Application secrets and environment configuration lived in AWS Secrets Manager and were consumed at runtime by Lambda handlers and EC2 services. CloudWatch logging covered authentication events, objects arriving in or changing within S3, and each push to Salesforce with its response, creating an end-to-end trace for the integration.
Integrating an enterprise client’s data
Protégitas’s largest enterprise client, a Mexican financial institution, delivered files of newly authorized policies. I replaced the previous infrastructure with an SFTP bridge using AWS Transfer Family, S3 and Lambda. The Lambda processed each delivered file and pushed its records into Salesforce over the REST API, ready for billing and member support.
The replacement removed manual steps and lowered the integration cost; neither outcome was quantified. The client’s identity and policy data remain private.
Taking platform ownership in house
On my own initiative, I took end-to-end ownership of Salesforce: the roadmap, requirement intake and prioritization, customizations and administration. I queried its data daily with SOQL for reporting.
This brought in house work previously performed by a consultancy billing thousands of dollars per month and reduced change-request delivery from days to hours. I continued to oversee the separate web development consultancy’s delivery; these were distinct responsibilities.
What this work demonstrates
My payments experience comes from delivering checkout and subscription integrations, tracing processor events and keeping operational systems reconciled. It also includes the product work around prioritization, insourcing and vendor delivery.
The independent payment-routing project is a separate technical demonstration using synthetic data. Its replay results are not outcomes from LegalBrave or Protégitas.