Senior ML Engineer: LLM/VLM Inference Optimizer

Jobgether SRL; Netherlands; On-site; EUR 120,000 - 180,000; Full time

Benefits offered by this job

Competitive compensation

Career growth

Ownership over technical work

International environment

Job summary

Jobgether SRL is seeking a Senior Machine Learning Engineer in the Netherlands to drive optimization of large language and vision-language model inference from artifacts to production. You will work across model internals, serving architectures, and benchmarking to improve latency and cost.

This hands-on role requires Python and PyTorch, experience with modern inference stacks, and the ability to design reproducible benchmarks and safe rollouts for high-throughput AI workloads.

Qualifications

• Strong Python and PyTorch software skills.
• Experience deploying or optimizing LLM/VLM inference systems.
• Experience with modern inference stacks.
• Understanding transformer bottlenecks including KV cache, attention, memory bandwidth.
• Ability to reason quantitatively about latency, throughput, and cost trade-offs.
• Experience diagnosing complex performance problems and translating findings into production improvements.
• Strong communication and collaboration skills.
• Experience with quantization/training optimization techniques is a plus.

Responsibilities

• Own optimization initiatives for specific model families and inference backends.
• Evaluate inference engines and recommend serving configurations.
• Diagnose and resolve production regressions.
• Optimize LLM/VLM endpoints for latency, throughput, memory use, GPU utilization, and cost per token.
• Deploy, benchmark, and extend inference engines (e.g., vLLM, Triton, TensorRT-LLM).
• Build model-compression workflows (quantization, distillation, low-bit serving).
• Implement advanced inference techniques (speculative decoding, KV-cache optimization).
• Develop reproducible benchmarkHarnesses covering latency and tokens/second per GPU.
• Collaborate with kernel/platform engineers to identify bottlenecks across stack.
• Investigate performance trade-offs and guide optimization with benchmarks.
• Produce design docs, performance reports, and rollout plans.

Skills

Python

PyTorch

LLM inference

VLM inference

Performance tuning

Benchmarking

System design

GPU/CUDA

Open-source

Collaboration

Tools

vLLM

TensorRT-LLM

Triton Inference Server

NVIDIA Dynamo

KServe

Ray Serve

SGLang

Job description

Jobgether SRL is seeking a Senior Machine Learning Engineer in the Netherlands to drive optimization of large language and vision-language model inference from artifacts to production. You will work across model internals, serving architectures, and benchmarking to improve latency and cost.

This hands-on role requires Python and PyTorch, experience with modern inference stacks, and the ability to design reproducible benchmarks and safe rollouts for high-throughput AI workloads.