System & Infrastructure Engineer

Il y a 3 heures

East Flanders, Flanders, Belgique Polysense Temps plein
h3Hybrid /h3 ul liGhent, Belgium /li /ul pPolysense builds operational intelligence software for food manufacturers. Companies like Agristo, Lotus Biscoff, and La Lorraine already run on it. /p pOur engineers build that on infrastructure we run ourselves: on-prem CPU and GPU servers, a basic Kubernetes cluster, our own CI and ML training runners, and cloud VMs on AWS and Azure. On top of that runs a lot of internal tooling, some built by us and some open source. /p pIt grew fast. Now it needs an owner. /p pThat's you. You own our infrastructure end to end, fix things properly instead of patching them, and lay the security foundations along the way. You report directly to the CTO/co-founder, and the learning curve is as steep as you want it to be. /p pMission: own Polysense's infrastructure end to end, so engineering stops firefighting and gets back to building. /p h3What You'll Be Doing /h3 ul liOwn the infrastructure, top priority: servers, CI and training runners, cloud VMs, and the internal tools – including spinning up extra compute when a big training run or GenAI experiment needs it. When something breaks, it's yours, and you make sure it never breaks the same way twice. /li liFrom firefighting to fixing: a training runner’s disk fills up and nobody can train models. A kernel update breaks GPU drivers. Right now both get a quick manual fix – you solve why they keep happening. /li liAutomate, AI-first: scripts, Terraform, Ansible, AI agents – repeatability and automation are your defaults for anything repetitive. You can fully offload eng-to-end tasks to agents, but also know when something is still too risky to hand off. /li liAccess, organized: GitHub and Azure access is managed by hand today. You build a clear role structure, so granting or revoking access takes minutes and nobody is left waiting. /li liSecurity foundations: you help build the real technical controls behind our ISO 27001 and SOC 2 path. You’re the builder, not the auditor. /li /ul pNot in scope: day-to-day laptop setup and new-hire onboarding. That's a pre‑configured IT pipeline, not yours. /p pKey point: we’re not looking for someone who’s seen it all. We’re looking for someone who wants to own it all, and learns fast doing it. /p h3What You Bring /h3 ul liA solid technical base: Linux, the command line, scripting in Bash or Python. You can debug and chase a root cause. /li liSome hands‑on experience keeping systems running. A first job, an internship, side projects, a homelab you’re slightly too proud of: it all counts. /li liExperience with, or real curiosity about, Docker, Kubernetes, cloud (AWS/Azure) and infrastructure‑as‑code like Ansible. /li liAn AI‑native way of working: setting up agents and automation to handle tasks is your default. /li liEarly career is fine: roughly 0–3 years of experience. /li liBonus points: exposure to ISO 27001 or SOC 2, a compliance platform like Vanta, GPU servers and Nvidia drivers, virtualization or storage setups, or networking basics. /li liNot required: senior‑level experience, certifications, or audit and legal compliance work. /li /ul h3Mindset way of working /h3 ul liOwnership by default: when something breaks, you don’t wait for someone to assign it to you. /li liHungry for the steep curve: you want a fast-paced environment where you learn something new every week. That energizes you, it doesn’t scare you. /li liFix it, then fix why: a band aid never quite satisfies you. You want the real fix. /li /ul pThere's real room to grow here: toward senior infrastructure or platform work, or deeper into security. We're not locking you into one lane. /p h3What Success Looks Like /h3 ul liMonth 1: you know the landscape. Every server, runner and VM, and where the docs are missing. /li liDay 60: the recurring fires, like full disks and broken GPU drivers, have a real fix instead of manual cleanup. /li liDay 90: engineering comes to you first with infra questions, and fewer of them reach engineering leadership. /li liGitHub and Azure access has a clear role structure. Granting or revoking it is quick. /li liYear 1: the firefighting load has clearly dropped, and you've got your own read on what to build or own next. /li /ul h3The Perks, What's in it for you? /h3 ul liReal ownership: it's your infrastructure, not a ticket queue. /li liA steep learning curve: GPU servers, Kubernetes, cloud and security in one role, with a team that backs you while you figure it out. /li liDirect line to the CTO/co-founder: no layers between you and the people deciding where this goes. /li liAmbitious, no-ego team: people aim high, dare to fail, and pick up whatever needs doing. /li liCompetitive salary in line with industry benchmarks, a full benefits package and a company car. /li /ul h3Why Polysense? /h3 pPeople join a company like Polysense for the impact they can have, the freedom to build something, and the growth, both theirs and ours. /p pWe're early-stage. We recently closed our €11M seed round and we’re