Vertex AI Architect

Role: Vertex AI architect

Location: Hyderabad

Shift Timings: 12:00 PM to 9:00 PM

Experience: 10+ Years

Roles and Responsibilities:

Solution Architecture: Design end-to-end AI/ML pipelines on Vertex AI, from data ingestion and model training to evaluation and model serving.

Hybrid Deployment Strategy: Architect and implement deployment pipelines that transition models developed in Vertex AI to on-premises, secure environments (utilizing technologies such as GKE Enterprise/Anthos or other container orchestration platforms).

MLOps Implementation: Standardize CI/CD/CT (Continuous Training) workflows to manage model versioning, monitoring, and automated retraining loops.

Infrastructure Optimization: Work closely with Infrastructure and Security teams to ensure on-premises environments are optimized for high-performance model serving, ensuring latency and throughput meet enterprise standards.

Model Portability: Develop strategies for containerizing models (Docker/Kubernetes) to ensure consistent behavior across Google Cloud and on-premises environments.

Cross-Functional Collaboration: Serve as a technical bridge between Data Scientists, Cloud Engineers, and On-Prem IT Ops to ensure seamless hand-offs and operational stability.

Required Qualifications:

Cloud Expertise: Proven experience with Google Cloud Platform, specifically the Vertex AI ecosystem (Vertex AI Pipelines, Feature Store, Model Registry, and Prediction services).

Hybrid Cloud/On-Prem Experience: Demonstrated success in deploying cloud-trained models to on-premises, air-gapped, or hybrid infrastructure.

Containerization Mastery: Expert-level knowledge of Docker and Kubernetes (GKE/Anthos/OpenShift).

AI/ML Foundations: Strong proficiency in Python, TensorFlow/PyTorch, and general machine learning engineering practices.

Infrastructure as Code (IaC): Hands-on experience with Terraform or similar tools for managing infrastructure across environments.

Security & Compliance: Experience working in secure environments; understanding of data sovereignty, encryption, and network security for on-premises deployments.

Preferred Skills:

Experience with Google Distributed Cloud (GDC) or hybrid cloud management tools.

Knowledge of model compression and optimization techniques (e.g., quantization, pruning) for local hosting constraints.

Professional certification in Google Cloud Professional Machine Learning Engineer or Professional Cloud Architect.