Production ML Engineer – NLP/Audio Models (Remote Germany)

Orcrist Technologies GmbH; United States; Remote; USD 130.000 - 170.000; Vollzeit

Benefits dieser Stelle

30 days vacation

Equipment & learning budget

Remote-first in Germany with Berlin-me

Zusammenfassung

Orcrist Technologies GmbH is seeking an ML Engineer to productionize NLP/audio/document models powering OIP. You will own packaging, deployment, monitoring, and evaluation in collaboration with Research and product squads.

You will work on streaming and batch inference, model evaluation pipelines, and cost/performance optimization on Kubernetes-based infrastructure, with remote-first collaboration across teams in Europe and the US.

Qualifikationen

• 4-8+ years ML engineering/MLOps experience shipping models to production.
• Strong Python and PyTorch/Transformers expertise.
• Experience with Triton/KServe or similar.
• Comfortable with Kubernetes, GitOps, CI/CD, and GPU workload operations.
• Knowledge of evaluation metrics, monitoring, and annotation workflows.

Aufgaben

• Package and deploy models (ASR, translation, OCR, NER, summarization) on Kubernetes using Triton/KServe.
• Build evaluation pipelines (WER, BLEU, F1, latency, cost) and automate release gating.
• Operate streaming + batch inference via Kafka, Temporal, and backfill tooling.
• Monitor drift/quality with Prometheus, Grafana, Evidently; optimize inference cost and performance.
• Collaborate with TypeScript teams on payload schemas, contracts, and human-in-the-loop feedback loops.

Kenntnisse

Python

PyTorch/Transformers

Kubernetes

GitOps

CI/CD

GPU workloads

Evaluation metrics

Monitoring

Annotation workflows

Tools

Triton

KServe

Prometheus

Grafana

Temporal

Beam

Flink

Ray Serve

ONNX/TensorRT

Jobbeschreibung

Orcrist Technologies GmbH is seeking an ML Engineer to productionize NLP/audio/document models powering OIP. You will own packaging, deployment, monitoring, and evaluation in collaboration with Research and product squads.

You will work on streaming and batch inference, model evaluation pipelines, and cost/performance optimization on Kubernetes-based infrastructure, with remote-first collaboration across teams in Europe and the US.