The problem
Two platform engineers were holding 40+ production workloads together at a fintech growing 3× year-over-year. An $85K/month bill was climbing with no reserved-instance strategy, off-hours incidents went undetected for hours, and an imminent SOC2 audit was already a sales blocker — promised to enterprise prospects without the controls to back it up.
What we shipped
We took over full managed operations: layered CloudWatch monitoring with business-level SLOs, a 15-minute critical-incident SLA, and AWS Config plus GuardDuty driving auto-remediation. We attacked cost with Compute Savings Plans, RDS right-sizing and dev-environment scheduling, then ran a 16-week SOC2 programme that formalised IAM federation, Terraform-only change management, Security Hub remediation and continuous evidence collection.
The outcome
Monthly spend fell from $85K to $61K — $288K annualised — even as transaction volume grew 35%. SOC2 Type II landed on the date sales had promised, unlocking three enterprise contracts worth over $2M ACV. Eight straight months of zero unplanned downtime followed, and the original two engineers shifted from 60% reactive ops to 90% strategic platform work.