Senior AI Machine Learning Engineer – Feature Engineering Delivery

Orbit Labs

Contract · Remote · Flexible hours

About the role

Orbit Labs is revamping the performance and reliability of its flagship customer‑facing analytics platform, codenamed "Nebula Engine". As an AI Machine Learning Engineer you will lead the redesign of model serving pipelines, ensuring low‑latency predictions and robust scaling under peak traffic.

Key responsibilities

• Refactor the existing TensorFlow Serving stack to a Kubernetes‑native MLOps workflow using Kubeflow Pipelines.
• Implement real‑time feature extraction in Spark Structured Streaming to feed models with sub‑second freshness.
• Profile and optimize model inference latency on GPU and CPU instances, targeting a 30% reduction.
• Design automated canary deployments with Istio traffic splitting and monitor SLA compliance.
• Collaborate with the data engineering team to integrate A/B testing results into continuous model retraining cycles.

Must-have skills

• Deep expertise in Python for ML model development and deployment.
• Strong experience with TensorFlow, PyTorch, and model serialization formats (ONNX, SavedModel).
• Proficiency in building and maintaining REST API endpoints for model inference.
• Hands‑on knowledge of Kubernetes, Docker, and CI/CD pipelines for ML workloads.
• Ability to profile and tune performance using tools like NVIDIA Nsight and Prometheus.

Nice to have

• Familiarity with MLOps platforms such as MLflow or Sagemaker.
• Experience with serverless inference on AWS Lambda or GCP Cloud Functions.