DevOps / TechOps Engineer
"I keep busy sites fast, cheap, and boring."
I run cloud and edge infrastructure for a portfolio of high-traffic publishing and fintech-adjacent web properties — the kind where a bad WAF rule costs traffic by lunchtime and a mis-sized task definition costs money all month.
Most of my work sits in three places: the edge, where Cloudflare rules decide who gets served and how fast; the pipeline, where changes ship through Terraform and GitHub Actions instead of a console; and the bill, where right-sizing and observability turn guesswork into a number someone can act on.
I like the unglamorous parts: least-privilege IAM, backup jobs that actually restore, migrations planned far enough ahead that nobody notices them happening.
AWS Certified Solutions Architect – Associate
Amazon Web Services Training and Certification
Verify on Credly →
AWS Certified SysOps Administrator – Associate
Amazon Web Services Training and Certification
Verify on Credly →
AWS Partner: Technical Accredited – Training
Amazon Web Services Training and Certification
Verify on Credly →
AWS Partner: Sales Accreditation
Amazon Web Services Training and Certification
Verify on Credly →
AWS Partner: Cloud Economics Essentials
Amazon Web Services Training and Certification
Verify on Credly →
Built a single pane of glass across a fragmented AWS estate: Cloud Cost Management rolled out org-wide via StackSets, log drains from the hosting platform, and frontend RUM instrumented with named business events — so product questions get answered without a ticket.
Prepared a high-traffic annual event: database engine migration and config porting, connection and thread-pool tuning, PHP-FPM pools moved from on-demand to static across five tenants, and nginx tuned per vhost. Handled live incident response during the window, including a connection exhaustion caused by one unbounded query.
Replaced blanket administrator access on shared automation roles with scoped policies derived from what the jobs actually do — database backups and storage sync across several projects — without breaking a single pipeline.
Tuned WAF and bot rules so the crawlers that matter get through and the ones that don't get dropped, tracked down a Worker bug returning stale responses to browsers, and rebuilt caching behaviour across the properties.
A Python service, deployed with Terraform on Lambda and scheduled with EventBridge, that pulls seat and usage data from several AI vendor admin APIs into one Notion workspace — so licence spend is visible without three logins.
My friend Njegos Vukadin makes AI portraits of an alternate-universe me.
AI portraits by Njegos Vukadin
Open to consulting, collaboration, or a conversation about edge caching that goes on too long.