ML Systems Engineer — Inference Acceleration
jobs.ashbyhq.com
- Required language
- English professional
- Job written in
- English
- Location
- Zürich
- Work type
- On-site
- Type
- Full-time
Arago is a deep‑tech startup that is reinventing computing by combining optical and CMOS technologies into a proprietary accelerator that delivers dramatically higher AI performance. Backed by prominent investors and industry leaders, including the CEO of Arm, the founder of macOS, an Nvidia Fellow, and Google's Head of Optics, Arago builds the only processor of its kind and expects its engineers to work at high velocity, own their projects, and collaborate as a single unit. The ML Systems Engineer will focus on optimizing the execution and serving of modern AI models on Arago's custom accelerator. Day‑to‑day tasks involve analysing AI workloads to spot kernel, runtime, memory and system bottlenecks; creating and refining custom kernels, fused operators and execution strategies; mapping models across multiple devices with efficient communication and synchronization; and implementing inference‑serving techniques such as continuous batching, paged KV caches and prefill/decode scheduling. The role also requires building profiling and benchmarking tools and partnering closely with hardware, compiler and runtime teams to co‑design software abstractions that influence future hardware capabilities. Candidates must bring strong experience in high‑performance ML inference, GPU or accelerator programming, and a deep grasp of computer architecture, accelerator execution models, memory hierarchies and parallelism. They need proven ability to develop and optimise custom kernels using environments like CUDA, Triton or ROCm/HIP, and to work with operator fusion, tiling, scheduling and data‑movement optimisation. Experience with distributed model execution, including tensor, pipeline, sequence or expert parallelism, and with modern inference‑serving stacks such as vLLM, SGLang or TensorRT‑LLM is required. Proficiency in C++ and Python is essential, and familiarity with MLIR or its dialects is a strong plus. The role demands fluent English. Nice‑to‑have skills include exposure to MLIR dialects and any prior work on custom accelerator stacks where compilers, runtimes and kernels evolve together. The position offers a competitive cash package, meaningful stock options, comprehensive healthcare, pension contributions, professional development support and 25 days of paid time off plus public holidays. Engineers will own a critical technical domain and have clear pathways for vertical or horizontal growth, driven by performance and personal ambition. What the role asks for: - Strong experience in high‑performance ML inference - Experience with GPU/accelerator programming - Deep understanding of computer architecture and memory hierarchies - Proven ability to develop custom kernels (CUDA, Triton, ROCm/HIP) - Experience with operator fusion, tiling and scheduling - Knowledge of distributed model execution and parallelism - Hands‑on work with inference‑serving systems (vLLM, SGLang, TensorRT‑LLM) - Proficient C++ and Python programming - English proficiency at a professional level - Nice‑to‑have: familiarity with MLIR and its dialects
Engineer: Lebenslauf-Vorlage
Bewirb dich mit einem Lebenslauf im Schweizer Aufbau — mit Beispieltext und den Anforderungen, die Engineer-Inserate am häufigsten nennen.
Lebenslauf-Vorlage Schweiz ansehenEngineer: was der Markt gerade verlangt
Engineer-Stellen gehören zu den regelmässig ausgeschriebenen Berufen auf SwissJobs.app.
Weitersuchen
- Alle Ingenieur Jobs in der Schweiz
- Alle Jobs in Zürich
- Lebenslauf als Engineer: Beispiel und Vorlage
- Lebenslauf mit KI erstellen
- ATS-Lebenslauf prüfen
- Bewerbungsschreiben für dieses Inserat