Work

Case studies

1. 300K+events / day1 week → 1 daypartner onboarding The Partner Layer Replaced a tightly coupled partner integration path with a standalone API service, cutting broadcaster onboarding from a week to a day at 300K+ events a day. Read the case study
   • TypeScript
   • NestJS
   • PostgreSQL
   • AWS SNS
   • Redis
2. 50+concurrent streams1 containerper live stream StreamForge A Go service that assembles a live HLS playlist by pulling from a broadcaster's asset manager, so they could onboard without buying their way into a format we already accepted. Read the case study
   • Go
   • AWS Fargate
   • AWS SNS
   • AWS S3
   • HLS
3. <100msp95 replication lag~300fields unpicked Moving a 300-field document to PostgreSQL, live Pulled clip variants out of an overloaded MongoDB document into normalized PostgreSQL models, replicated off the existing event stream, with no downtime and no cutover date to hit. Read the case study
   • MongoDB
   • PostgreSQL
   • Go
   • AWS SNS
   • AWS EC2
   • TypeScript