Senior AI Inference Engineer — Kernel & Edge Optimization

Lever, Inc.; Schweiz; Remote; EUR 137.000 - 222.000; Vollzeit

Benefits dieser Stelle

Remote-first team

International team

Cutting-edge AI research

Performance-critical infra

Collaboration with researchers

Edge deployment challenges

Zusammenfassung

Lever, Inc. is seeking an AI Research Engineer (Kernel & Inference Optimization) based in Switzerland. You will push the boundaries of efficient AI systems across hardware environments, balancing latency, throughput, and memory in model-serving architectures.

You will research and implement advanced inference strategies, develop custom GPU kernels, and collaborate across teams to translate research into production-ready infrastructure for edge devices.

Qualifikationen

• PhD in NLP/ML with strong AI research track record and publications.
• Proven expertise in MSL and writing custom compute shaders.
• Experience with low-level kernel optimization on mobile/resource-constrained devices.
• Track record delivering measurable improvements in latency, throughput and memory.
• Deep understanding of model-serving architectures and optimization techniques.
• Experience writing GPU kernels for mobile devices.
• End-to-end inference pipelines from model optimization to production on constrained hardware.
• Ability to apply empirical research and systematic experimentation to solve bottlenecks.
• Experience benchmarking inference systems across scales and environments.
• Knowledge of distributed inference: tensor/pipeline/expert parallelism for large GPUs.
• Understanding diffusion models and Vision Transformers architecture.
• Familiarity with pruning, quantization, Flash Attention, KV Cache, speculative decoding.

Aufgaben

• Design and deploy advanced model-serving architectures for high throughput and low latency.
• Develop inference pipelines for mobile and edge platforms with constrained resources.
• Set performance targets for latency, throughput, memory and reliability.
• Build controlled benchmarks and collect latency, throughput and memory metrics.
• Create datasets and simulations to evaluate model performance in real-world conditions.
• Identify bottlenecks and implement solutions in batching, networking and memory management.
• Develop custom GPU kernels and compute shaders for mobile hardware (MSL).
• Apply inference optimization techniques including pruning, quantization and Flash Attention.
• Design distributed inference with tensor/pipeline/expert parallelism for large GPU workloads.
• Collaborate with cross-functional teams to integrate optimized inference into production and edge apps.
• Document results and refine optimization strategies based on empirical evidence.
• Monitor production performance and seek improvements in scalability and efficiency.

Kenntnisse

Latency optimization

Benchmarking

Empirical research

English communication

Tensor parallelism

Distributed inference

Ausbildung

PhD in NLP or ML

CS degree

Tools

Metal Shading Language (MSL)

Jobbeschreibung

Lever, Inc. is seeking an AI Research Engineer (Kernel & Inference Optimization) based in Switzerland. You will push the boundaries of efficient AI systems across hardware environments, balancing latency, throughput, and memory in model-serving architectures.

You will research and implement advanced inference strategies, develop custom GPU kernels, and collaborate across teams to translate research into production-ready infrastructure for edge devices.