Helped scale a high-growth social-commerce platform through explosive, celebrity-driven growth, leading an all-in cloud migration that absorbed massive, unpredictable traffic spikes at controlled cost. The platform reached national prominence, and its cloud economics were later documented in a public cloud case study.
Context and stakes
I moved from senior engineer to the platform's principal engineer and de-facto technical lead, reporting directly to the founders, on a fast-growing social-commerce platform with a rewards system used across a large merchant network, as the engineering organization scaled from 3 to 300+. Growth was volatile and often celebrity-driven, producing unpredictable traffic spikes that the platform's fixed-capacity infrastructure could not absorb.
Problem
Infrastructure limits were causing instability before migration, including outages during growth periods. The platform needed to move from fixed-capacity constraints to an architecture that could absorb rapid demand changes without runaway cost.
Intervention
I helped lead an all-in cloud migration onto managed services (relational, serverless, CDN, container orchestration, and media processing), shifting the platform away from brittle, self-limited infrastructure. I also provided foundational engineering behind the rewards transaction service used across multiple product surfaces.
Outcome
- #3 most-downloaded app in the Philippines (January 2021), per Data.ai via Rest of World.
- 30M+ users at peak.
- 3M daily active users in publicly reported windows.
- An AWS case study documented chat capacity scaling from 100K to 1M simultaneous users.
- The same case study documented about +600% user growth with only about +50% monthly operating-cost increase.
References
- AWS public cloud case study (archived): view the archived case study
- App ranking, Data.ai via Rest of World: restofworld.org
Reusable lesson
Growth only counts if the unit economics hold as you grow. Managed cloud adoption removed hard capacity ceilings here, but the result that matters is absorbing rapid user growth at only a fraction more monthly cost. Engineer for cost-efficient scale, not just peak capacity.