About the Role

You will take customer language models into production on our GPUs or the customer's, design serving stacks that are performant, reliable and cost-efficient, and operate them after launch.

Responsibilities

Deploy open-weight models on vLLM, TensorRT-LLM or Triton.
Tune batching, quantization and caching to meet latency and throughput targets.
Develop pipelines for RAG, evaluation and model releases.
Implement API gateways with authentication, rate limiting and usage tracking.
Build observability covering latency, errors, token usage and cost per request.
Plan capacity with the GPU infrastructure team.

Qualifications

At least 3 years of experience in DevOps, platform engineering or MLOps.
Strong proficiency with Kubernetes, Helm and containers.
Strong Python skills and the ability to read inference-server code.
Experience deploying LLMs or deep learning models to production.
Experience with Prometheus, Grafana or OpenTelemetry.

Preferred Qualifications

Understanding of KV cache, speculative decoding or tensor parallelism.
Experience with vector databases such as Milvus, Qdrant or pgvector.
Open source contributions to LLM tooling.

Tools and Technologies

Kubernetes
Helm
vLLM
TensorRT-LLM
Triton
Python
Prometheus
OpenTelemetry

Working at Zotect

Work across the full AI infrastructure stack, from data centers, networking and GPUs to model serving.
Deliver production systems for customers in Thailand, Singapore and the United States.
Work closely with leadership in an engineering-led company.
Our office is located in the Gaysorn Building on Ploenchit Road, next to BTS Chit Lom.

How to Apply

To apply, send your CV and any relevant portfolio links to hr@zotect.com with the position title in the subject line.

Please include only the information needed to assess your application and omit sensitive data such as national ID numbers, religion or health information. For details, see our Privacy Policy