Location: US Remote
Candidates must currently live in the US and have work authorization to work
for any US employer
About Friday Harbor:
Friday Harbor is the leading AI-native technology platform in US mortgage, the
largest sector of consumer credit in the world.
We’re already used on almost 1% of all residential mortgages in the US, and that
number is growing rapidly. Three of the top 25 lenders in the country trust
Friday Harbor in production today, which means the work you ship here touches
real loans and real borrowers right away.
We’re a Seattle-based startup backed by leading venture firms, including
Abstract, Mischief, Wischoff, AI2 Incubator, and CoFound.
Our engineering team is small, fully US remote, and comes together for regular
off-sites. We care about doing ambitious work while offering flexibility and
good work-life balance.
The opportunity:
How do you reliably run AI agents in production? How do you schedule their work,
and ensure they are allocated sufficient resources? How do you understand what
they’re doing, recover when they fail, and keep costs predictable as usage
grows?
We’re looking for a Senior Infrastructure Engineer to work closely with our Lead
AI Platform & Infrastructure Engineer on the foundation for our next stage of
growth. You’ll help decide how our infrastructure evolves, lead the work to get
it there, and own the systems you put into production.
There’s meaningful room to design here, alongside a live product and real
customers to support. You’ll work with engineers who have infrastructure and
platform backgrounds and care deeply about getting these decisions right.
We want a strong managed infrastructure foundation, with clear abstractions that
let our team build on top of it. That takes someone who understands the
underlying systems well enough to know where we need control, where a managed
service makes sense, and when the simplest approach will serve us best.
What You'll Own:
Evolve our production infrastructure. Help define the architecture and lead
incremental migrations toward a more flexible, cloud-native deployment model.
Make changes safely, with clear rollout and recovery plans.
Build the foundations for running AI workloads. Work through questions around
execution environments, workload isolation, scheduling, orchestration, and
resource utilization. Evaluate new approaches against concrete product needs.
Make production understandable and reliable. Build observability that helps
us understand system behavior, diagnose failures, and make informed decisions
about performance, capacity, and cost.
Improve how engineers ship. Develop infrastructure as code, deployment
workflows, and useful defaults that make it easier to build, test, and
operate services.
Own the operational follow-through. Participate in on-call, improve recovery
procedures, test backups, and turn incidents into lasting improvements.
Build security into everyday engineering. Make access controls, secrets
management, patching, and data protection part of how our infrastructure
works.
Shape the boundary between infrastructure and platform. Partner with our
engineering and biz-ops teams to understand what they need and turn those
needs into capabilities they can depend on.
What We're Looking For:
Deep Kubernetes experience in production. You’ve operated Kubernetes for
substantial workloads and have a strong mental model of scheduling,
networking, resource management, scaling, and failure modes. You can reason
through what’s happening when a deployment or cluster behaves unexpectedly.
Experience designing new infrastructure and migrating existing workloads onto
it. You’ve made architectural decisions with incomplete information, moved
beyond deployment patterns a team had outgrown, and carried the work through
to production.
Practical engineering judgment. You weigh the value of additional control
against the complexity and ongoing work it introduces. You can explain when
you chose a managed service versus when you built something yourself.
Strong cloud fundamentals. You’re comfortable with modern cloud platforms
(AWS preferred, or Azure, Google Cloud etc.), Linux, networking, storage, and
infrastructure as code such as Terraform. You understand how these pieces
interact when troubleshooting production systems.
The ability to build and automate. You can write software and tooling that
removes operational work and makes infrastructure easier for other engineers
to use.
Ownership and clear communication. You’re comfortable taking an ambiguous
problem from investigation through implementation and operation. You explain
tradeoffs clearly and bring other people into decisions that affect their
work.
We expect this experience will usually come from five or more years in
infrastructure, platform, SRE, or a related engineering role. The systems you’ve
owned and the judgment you’ve developed matter more than an exact number of
years.
Nice to Have:
Operating AI applications, agentic workloads, or distributed systems.
Building execution environments with strong isolation or working with
serverless compute.
Supporting Python services and asynchronous workers.
Working on a small team where you helped establish infrastructure practices.
Operating systems that handle sensitive data or support regulated customers.
You don’t need to have experience in every area before joining. Strong
infrastructure fundamentals and curiosity about these problems are more
important to us than the specific technology you’ve used.
Why Join:
You’ll have substantial ownership of decisions that shape how Friday Harbor
builds and operates. The work is close to the product, has real customer impact,
and there are many interesting challenges left to solve.
If you enjoy getting deep into how systems work, making thoughtful tradeoffs,
and building something a small team can confidently operate, we’d like to talk.
We offer a base salary of $205,000 - 225,000 equity, and a comprehensive
benefits package.