Guilherme
Jaccoud

Own platform reliability for a multi-region streaming network serving 90M MAUs. Lead a 22-person SRE org across edge, control plane, and data infra.

• Designed a cell-based architecture rollout that contained 11 control-plane incidents to single cells — zero customer-facing outages in 18 months.
• Rewrote the autoscaling controller in Rust; reduced p99 scale-up latency from 38s to 4.2s and cut compute spend by $11M/yr.
• Established the org's first error-budget policy and quarterly reliability review; adopted by 9 product teams.
• Mentored 6 senior engineers to staff and 2 staff engineers to principal.

Led the 6-person SRE team through Talabat's post-acquisition migration from an on-prem monolith to a cloud-native distributed system on AWS; owned all infrastructure from networking to secrets.

• Built greenfield AWS infrastructure from scratch — VPC, networking, and EKS — using Terraform modules co-authored with the Security team; migrated SCM from Azure DevOps to GitHub and deployed Atlantis for self-service infrastructure changes.
• Implemented AWS Direct Connect between the on-prem DC and AWS to enable gradual, zero-disruption service migration; retained the primary database on-prem post-migration where DC latency outperformed RDS.
• Partnered with development teams on the migration from legacy .NET Framework to .NET Core on Kubernetes; introduced the Serverless Framework for workloads where dedicated Kubernetes deployments were disproportionate.
• Built a GitOps CLI orchestrating deployments from service repositories to an infra monorepo, integrated with ArgoCD; delivered multi-region active-active EKS clusters with 50/50 traffic split via Cloudflare Load Balancer and automatic failover.
• Owned the full platform stack: secret management with HashiCorp Vault, policy enforcement with OPA Gatekeeper, and regular Chaos Monkey exercises to validate disaster recovery scenarios.
• Designed a daily load-testing pipeline — Terraform + K6 + K3s — triggered by a scheduled GitHub Actions workflow at 4 AM; spun up ephemeral infra, ran developer-authored test suites against production, and streamed all metrics to New Relic, providing daily visibility into scalability headroom.
• Wrote a Rust API + React UI for engineer onboarding that automatically provisioned credentials across internal systems and third-party vendors, eliminating manual setup steps.

Infrastructure hire following Symphony's acquisition by Google; led the platform re-architecture from a monolith to microservices on GCP, with per-tenant deployments across dedicated cloud accounts for each enterprise client.

• Drove the monolith-to-microservices decomposition, migrating all services onto Kubernetes on GCP as part of the post-acquisition re-platform.
• Designed a per-tenant deployment model where each enterprise client ran their own fully isolated Symphony stack across dedicated GCP and AWS accounts — provisioning and maintaining dozens of independent environments in parallel.
• Led IaC migration from ad-hoc Python scripts + CloudFormation to multi-cloud Terraform (GCP/AWS); authored reusable modules covering networking, persistence, and orchestration layers.
• Built a GitOps CLI in Go that resolves module dependency order (networking → persistence → orchestration), applies Terraform, and commits the resulting state to Git — making every environment change auditable and reproducible.
• Integrated the CLI into Jenkins, giving development teams a self-service UI to spin up, update, and tear down development and production environments without SRE intervention.
• Operated a heterogeneous persistence tier — HBase, Hadoop, MongoDB, Solr, Elasticsearch, Hazelcast — on auto-scaled GCP instances across all tenant environments.
• Trained directly by Google Cloud engineers and reported to them for all infrastructure architectural decisions throughout the engagement.

Founded Latin America's first managed WordPress hosting company; grew it to thousands of customers and a team of 12, then sold the business in 2016.

• Designed per-tenant isolation on AWS — each customer site provisioned with its own Auto Scaling Group, ALB, and Cloudflare WAF + CDN — guaranteeing resource isolation across thousands of hosted WordPress properties.
• Rebuilt the entire platform on Kubernetes in 2014 — among the earliest production deployments in Latin America — reducing environment provisioning from hours to minutes and enabling zero-downtime rolling deploys.
• Architected the application stack: NGINX / PHP-FPM / MariaDB / Varnish, with static assets on S3; later migrated to managed RDS and Redis full-page cache, improving hit rate and eliminating a class of database failover incidents.
• Sole on-call operator for the first two years; grew the engineering and operations team from 0 to 12, owning all hiring, incident response, and capacity planning through to the 2016 exit.