Platform & Site Reliability Engineer
Mentimeter · Stockholm · Jun 2022 – now
Keeping an interactive presentation platform healthy for a live audience, where a slow response is as visible as an outage.
- Led the migration of production workloads from Heroku to AWS ECS, improving application performance by 20% and reducing hosting costs by ~10%, while giving engineers a platform they could experiment on freely.
- Built and operated a multi-account AWS structure with centralised identity through IAM Identity Center, giving teams isolated environments with consistent access controls and guardrails.
- Worked with product engineers to establish a common OpenTelemetry tracing architecture across products and teams, giving consistent request visibility from frontend through backend services.
- Reduced on-call toil by auditing 90 days of Datadog monitors, then building Cursor-based agents that assess incoming alarms and attribute ownership, cutting manual triage effort by ~50%.
- Migrated the majority of infrastructure to Infrastructure-as-Code (CDKTF/Terraform), replacing manual console changes with reviewed, version-controlled deployments.
AWSDatadogTerraformOpenTelemetryOIDC / IAMSLOs