Fig. 01
Backend design
Services that survive traffic spikes and flaky third-party APIs: queues, background processing, idempotency. Strict typing and tests as the norm, not a pre-release chore.
InfrastructureBackendSecurity
Systems that keep working when nobody is watching
I design and operate server infrastructure and Python backends. Fifteen years of owning systems in production — built to hold up in operation, not just to pass acceptance testing.
01 — Work
What I design and operate
Fig. 02
Infrastructure & operations
Deployment, containers, networking, observability. A backup counts as working only after a verified restore — everything else is hope.
Fig. 03
Security
Access control, encryption, service isolation, minimal attack surface. Designed in from day one: bolting it on later usually means never.
Fig. 04
Technical leadership
Reviews, engineering standards, automated checks in CI. Quality held by process and tooling, not by individual heroics.
02 — Approach
Four rules I work by
01
Understand first
No solutions for systems I haven't read. The worst architectural decisions are made before looking at how things actually work.
02
Simple wins
No abstractions for a hypothetical future: the layer added “for growth” is usually the one that blocks it.
03
Verify, don't trust
Working is proven by running; recovery is proven by recovering. A green status report is not evidence.
04
Automate the repeatable
Anything done by hand twice becomes a script or a check. It's about reproducibility, not saved minutes.
03 — Proof
Infrastructure you can touch
This site, the domain's mail, DNS and monitoring run on infrastructure I designed and operate myself. Every decision survives my own production first — only then do I recommend it to anyone else.
- Full cycle: architecture → development → deployment → operations
- 24/7 services with monitoring and alerting
- CI with type checks, tests and dependency audits
- Encrypted backups with verified restores
15
years in engineering
24/7
service operation
Full
cycle of work
04 — Contact
Let's discuss your project
A new system, an audit of an existing one, or ongoing operations. Describe the task — I'll tell you how I can help, and I'll say honestly if I can't.