Principal Engineering & R&D
The gnarly stuff: distributed systems, performance at scale, network & infra automation, novel prototypes taken to production. Brought in when the problem is genuinely hard.
- architecture
- performance
- prototyping
Remote • GlobalDeep engineering
Deep engineering
for the problems you
can't staff for.
Principal-grade software engineering, R&D, and fractional technical leadership — focused on generative AI, agentic orchestration & infrastructure, and AI-assisted development, built on deep roots in HPC, networking, cloud and product engineering. I build the foundations your team scales on, then make sure they can excel without me.
quadzero — zsh
$ whoami
principal engineer · oss maintainer
$quadzero --bind 0.0.0.0
distributed systems · network & infra automation · performance & scale · team & architecture leadership
ready listening on all interfaces
20+
years in production systems
4
continents of clients served
8
deep-expertise domains
OSS
maintainer & contributor
What I do
Three ways to put a principal engineer on your hardest work.
Fractional Eng Leadership
Fractional CTO / VP / principal for startups and scaleups. I set technical direction, build the foundations for scale, level up the team — and hand it back running.
- tech strategy
- scaling
- mentorship
Open Source & Advisory
OSS strategy from someone who maintains it: upstream contribution, building in the open, community, plus technical due diligence and architecture review.
- oss strategy
- due diligence
- review
Where I go deep
Domains, not just code.
Real context in the industries where the engineering is hardest — earned across operators, vendors and research.
Agentic Systems & Infrastructure
Enterprise AI & HPC
Cloud Infrastructure
Data Center & Telecom Networking
Software Security
Computational Sciences
Health Tech
EdTech
Track record
Problems solved across operators, vendors & research.
Through direct engagements and prior roles. Names shown for context — not active or implied endorsements.
Selected work
A look at the shape of the work.
Two representative engagements — problem, approach, and what the team was left running.
Data-Center & Telecom Networking
Cutting path-setup latency for a global interconnect fabric.
Re-architected the control plane behind on-demand cross-connects so circuits provision in seconds, not minutes — without a forklift upgrade.
−61%p99 provisioning time
14 metroszero-downtime rollout
// Case study coming soon
Enterprise AI & HPC
A self-serve inference platform on bare-metal GPU.
Took a research cluster that only its authors could run and turned it into a scheduled, multi-tenant platform the whole org ships models on.
3.4×throughput per GPU-hour
−45%cost per 1M tokens
// Case study coming soon
How I work
A senior partner, not a vendor portal.
Built to be left
Foundations your team can own and run after I'm gone. No lock-in to me.
Opinionated & honest
I'll tell you the trade-off and the thing you don't want to hear. Impartial by default.
Time-zone native
Estonia-based, structured to work across US, AU, UK and EU. Async-first, on all interfaces.
Contact
Got a problem worth a principal engineer?
Tell me what you're stuck on. If it's a fit, we'll scope it on a call. If it isn't, I'll point you to someone better.