You sit between our developers, our researchers and the machines they run on, from GPU servers and a compute cluster to trading servers in exchange colocations. A small team covers all of it: every host, service and secret is defined in code, and when something breaks, the alert lands in front of you, not in a support ticket queue. We are also building the platform our AI coding agents run on, and it is a frontier where few companies our size are working. Mistakes here have real and immediate consequences. So does good work.
DevOps Engineer
Our code trades on four continents around the clock. Your infrastructure is why it can.
Position
DevOps Engineer
Employment
Full-time
Work Model
On-site
Location
Spork Palace, Prague 1
What You Will Do
- Own our infrastructure as code: Ansible roles, rootless Podman services, and secrets in Vault, from study servers to colocation hosts.
- Build and run observability: Grafana, Prometheus, Loki, and AlertHub, our own alerting service. Alerts should page the right person at the right hour, and only when it matters.
- Run the developer platform everyone depends on: GitLab and its CI runners, Gerrit, Buildbot, shared Python environments, and GPU and cluster access.
- Build safe infrastructure for AI agents: sandboxed runners with controlled network access, autonomous coding agents, and MCP servers that connect them to our tools.
- Plan and carry out migrations without anyone noticing: storage systems, new servers, decommissions, OS and kernel upgrades.
- Automate repetitive work, then automate whatever's left.
Most of our day is thinking, arguing toward the best solution, and coding. We like results reasonably fast, in both development and runtime, and we give people real room to run and a direct line to the result.
Who We Are
Qminers is a Prague-based algorithmic trading firm competing on four continents, 24/7. Founded in 2012, we delivered a Sharpe ratio above 10 year after year for a decade. We are a team of around 50: mathematicians, quants, software engineers, and machine learning researchers.
Excellence is not a slogan here, it is the baseline.
Requirements
Must have
- Fluency with Linux and the shell, and a real understanding of what happens underneath: processes, filesystems, networking, systemd.
- Hands-on experience with configuration management (Ansible or similar) and containers in production.
- Good command of Python.
- Excellent logical thinking, and a habit of finding the root cause instead of restarting and hoping.
- Ownership: you'd rather be woken by an alert you set up than surprised by one you didn't.
Nice to have
- Experience running databases, message queues or other stateful services, not just deploying them.
- Prometheus, Grafana, Loki or similar observability tooling, especially alerting rules kept in code.
- GitLab CI, HashiCorp Vault, Kafka, PostgreSQL tuning.
- Experience keeping live, latency-sensitive systems running, not just staging environments.
- Interest in sandboxing and securing AI agents, or experience with it.
- Experience with storage at scale (NFS, parallel filesystems), GPUs or HPC schedulers.
- A track record of making other people's work faster or easier, not just your own.
We are honest about fit. This is not a demo and it is not a benchmark: the data is genuinely hard, the signal is faint, and a result only counts when it holds up on real metrics. If you want to be only a researcher who hands off, or only an engineer who ships someone else’s model, this is the wrong seat. If you want to be both, it is the right one.
Our Culture
The best argument wins.
Meritocracy is in our DNA.
You own the outcome.
Responsibility here is total, the wins and the mistakes alike.
Your rivals are outside, not inside.
Excellent colleagues are the elevator up, not the competition.
Interview process
01
Application
Through the web form.
02
Screening call
A short conversation to align expectations. You know where you stand.
03
Technical screening
An online test and a short technical call.
04
Take-home task
A real problem, a reasonable amount of time, and detailed feedback either way.
05
On-site interview
One-to-one conversations with the team, and lunch together.
06
Decision within a week
A clear answer, and if it did not work out, what was missing so you can try again.
Compensation
Our package has four parts.
Base salary
Every role has a published salary range, and the more experience and results you bring, the higher you move within it.Profit share & bonuses
Everyone holds a share of net profit, up to low single-digit percentages depending on impact. Around a quarter of the firm’s total annual profit goes into bonuses, re-evaluated every year on real, measurable impact.Shadow equity
A meaningful share in the firm’s performance, a standard part of compensation for senior contributors.ESOP
For the most senior people, genuine ownership in what the firm becomes. The details are worth a conversation.We publish our annual results, and anyone inside the firm can see how Qminers is paid and how the current year is shaping up. Financial transparency is a given, not a perk.
Interested in working for Qminers?
DevOps Engineer