Software Engineer (LLMOps)

Brak informacji o wynagrodzeniu

MidZdalnie·ManisesFull-time·Umowa o pracę

#471063·Dodano 5 dni temu·0

Źródło: justjoin.it

Aplikuj teraz

Tech Stack / Keywords

vLLMSGLangGPU infrastructureKubernetes Docker Terraform CI/CD

Podsumowanie oferty

Build and optimise LLM serving infrastructure for enterprise AI, including GPU infrastructure, model governance, and automated deployments. Work with inference engines such as vLLM and SGLang, alongside Kubernetes, Docker, Terraform, and CI/CD.

Wymagania

Strong background in Infrastructure Engineering, Platform Engineering, MLOps, or LLMOps, with experience running ML/LLM workloads in production.
Hands-on experience with LLM serving technologies such as vLLM, SGLang, or similar.
Practical knowledge of GPU infrastructure, orchestration, scaling, and inference optimisation.
Strong production experience with Kubernetes, Docker, Terraform, and CI/CD.
Proficiency in Python and/or Go.
Experience with monitoring, observability, and troubleshooting production systems.
Understanding of inference cost, latency, throughput, and performance trade-offs.
Strong ownership mindset and ability to collaborate across engineering teams.

Nice to have:

Experience with on-premise, air-gapped, or regulated enterprise environments.
Familiarity with LiteLLM, Langfuse, Prometheus, Grafana, or similar tools.
Experience with model registries, prompt versioning, and LLM governance.
Knowledge of distributed inference and advanced GPU optimisation.

Obowiązki

Build and optimise LLM serving infrastructure using technologies such as vLLM, SGLang, or equivalent inference engines.
Manage GPU infrastructure, including orchestration, scaling, load balancing, capacity planning, and high availability.
Optimise inference performance and costs through batching, quantisation, caching, and GPU utilisation improvements.
Build monitoring and observability for model serving, covering infrastructure health, latency, errors, and model output degradation.
Support enterprise deployments into on-premise, private, and air-gapped environments, ensuring security and reliability.
Develop model governance infrastructure, including model registries, versioning, LLM gateways, logging, and auditing.
Build and maintain CI/CD pipelines and infrastructure-as-code for automated, reliable model deployments.
Troubleshoot production issues and collaborate with Applied AI, Foundational AI, and SRE/Platform teams.

Maisa AI

6 aktywnych ofert

Zobacz wszystkie oferty

Aplikuj teraz