About DNAstack At DNAstack, our mission is to power precision medicine by building software that breaks down barriers to responsible biomedical data sharing, discovery, and analysis. We build cutting-edge software and…
Compensation: £149k – £213k Qumulo's cloud data platform manages exabytes of the world's most demanding data, unifying files, objects, and every workload across edge, core, and cloud. About the Role Are you a…
Compensation: $254k – $295k Qumulo's cloud data platform manages exabytes of the world's most demanding data, unifying files, objects, and every workload across edge, core, and cloud. About the Position The…
Le code de simulation code_saturne (logiciel open source disponible sur www.code-saturne.org) capitalise depuis plus de 25 ans le savoir-faire des équipes EDF dans les domaines de la thermo-hydraulique et de la CFD…
Help Build the Future of Enterprise AI Infrastructure Enterprise AI is changing how organizations think about technology infrastructure, and Cerio is helping shape what comes next. We’re building an open systems…
AI Engineer focused on GPU and LLM optimization, accelerating inference and serving pipelines with frameworks like vLLM and TensorRT-LLM, writing CUDA/Triton kernels, and deploying low-latency AI endpoints on Kubernetes for a hybrid compute platform, with remote options in Canada, UAE, or worldwide.
Umbra is an American space technology company delivering advanced systems, from sensors to spacecraft, that empower customers worldwide with unmatched access to critical information from space. Our mission is simple…
Senior HPC Systems Engineer supporting Nscale's GPU cloud in APAC. Acts as a senior escalation point diagnosing GPU node, InfiniBand/RoCE fabric, and Linux infrastructure faults, managing Slurm HPC jobs, and mentoring engineers while participating in on-call rotations and occasional onsite travel.
Part of Nscale's APAC support rotation, the engineer handles tickets and alerts, triages GPU nodes and server hardware (nvidia-smi/DCGM, BMC, mlxlink), runs fabric and storage diagnostics, follows runbooks, and keeps NetBox/DCIM records accurate, with on-call duties and occasional travel. Core stack: GPU/HPC infrastructure, Linux, networking, Bash/Python, Git.
Overview DDN is seeking a Technical Account Manager (TAM) to work directly with strategic enterprise and AI/ML customers, deploying, integrating, and optimizing DDN systems in production environments. As an TAM, you…
Lambda, The Superintelligence Cloud, is a leader in AI cloud infrastructure serving tens of thousands of customers. Our customers range from AI researchers to enterprises and hyperscalers. Lambda's mission is to…
As a Cybersecurity Engineer, you bring deep expertise in incident response, detection engineering and forensics across Windows/Linux/cloud environments/network. You hold certifications like Certified Incident Handler,…
As an Detection Engineer, you you bring solid software development skills in either Rust, C, C++ or similar and deep knowledge of operating systems (Linux & Windows), ideally with a proven background in building…
Compensation: $120k – $200k Senior Technical Marketing Engineer Remote, U.S. (Travel up to 25%) Reports to: Lab Lead, FarmGPU AI Lab Employment status: Full-time About FarmGPU FarmGPU operates bare-metal NVIDIA GPU…
Compensation: $235k – $265k Qumulo's cloud data platform manages exabytes of the world's most demanding data, unifying files, objects, and every workload across edge, core, and cloud. About the role: This is a…
Compensation: $260k – $270k Qumulo's cloud data platform manages exabytes of the world's most demanding data, unifying files, objects, and every workload across edge, core, and cloud. About the Position: This…
Compensation: $250k – $275k Qumulo's cloud data platform manages exabytes of the world's most demanding data, unifying files, objects, and every workload across edge, core, and cloud. About the role: This is a…
Principal Systems Engineer owning core components of kernel test and validation, measuring and optimizing platform performance, and contributing to key system components at the OS/kernel/GPU/CPU level. Works on automated GPU checkpointing infrastructure for AI+HPC clusters.
Compensation: $250k – $300k Founding Head of AI Infrastructure Spacebus | Hire 01 | Full-time | Reports to: Founder | Regular site travel The opportunity Spacebus is at the 0–1 stage of building a distributed AI…
Cookies?
A session cookie keeps you signed in. If you accept, we also use analytics cookies (Google Analytics and PostHog) to understand usage. Learn more
We couldn't check your fit for this role — add a CV to your profile to see it next time.