A
MLOps Architect/Real-Time Inference Lead
External job listingat Arkhyatech
MLOps Architect / Real-Time Inference Lead in Charlotte, NC. Contract, onsite. Send your updated resume by email or comment/DM on LinkedIn.
External source - not verified5 weeks agoOpen until: Jan 13, 2027
Salary
Not provided
Location
Charlotte, United States
Employment type
Full time
Workplace
Not provided
MLOps Architect/Real-Time Inference Lead
Charlotte, United States
Job description
Tasks
- Design, build, and industrialize low-latency, highly available real-time model serving platforms (sub-second SLAs).
- Establish production deployment strategies (Canary, Blue/Green, Shadow deployments).
- Implement custom Kubernetes autoscaling (HPA, custom metrics) to handle fluctuating traffic and optimize cloud spend.
- Perform performance profiling, load testing, and latency optimizations (quantization, GPU/TPU acceleration, pruning).
- Build CI/CD & GitOps pipelines for automated model testing, packaging, and validation.
- Implement robust observability for tracking model drift, SLOs/SLIs, and system health.
Requirements
- Deep hands-on experience with Triton Inference Server, TorchServe, TF Serving, vLLM, Seldon Core, or KServe.
- Advanced experience with K8s, HPA, microservices, and hybrid cloud/on-prem deployments.
- High proficiency in gRPC, REST, or GraphQL for online inference.
- Hands-on with load testing tools (Locust, JMeter, K6) and model optimization techniques.
- Prometheus, Grafana, Datadog, ArgoCD, GitHub Actions, or Jenkins.
- Strong Python, Go, or C++, plus PyTorch / TensorFlow / Scikit-Learn frameworks.
Schedule
- Contract
- Onsite
How to apply
- Send your updated resume directly by email.
- Or drop a comment/DM on LinkedIn for immediate review.
Location
- Charlotte, NC