Gen AI Architect

Title : GEN AI Architect

Location - NJ - onsite

Job Description

Responsibilities :

Lead the architectural design and implementation of scalable, generative AI-powered solutions for business automation, leveraging both Azure and Google Cloud platforms.

Architect end-to-end AI model workflows, utilizing Azure AI, and Google AI tools. Focus will be on applications involving complex document intelligence, multi-modal data analysis, and advanced image processing tasks.

Design and oversee the development of optimized RAG (Retrieval Augmented Generation) pipelines that incorporate varied data sources and formats to improve prompt execution.

Establish best practices for prompt engineering and iterative refinement to maximize the accuracy and relevance of outputs across diverse data types.

Collaborate with cross-functional teams (engineering, data science, business) to translate complex business requirements into robust, scalable AI component designs.

Champion the integration of the client's proprietary generative AI tools into the architecture and development lifecycles.

Direct the implementation of robust API management strategies using Apigee for secure and efficient access to AI models and related services.

Ensure all solution designs adhere to stringent data governance policies and use cloud-native security mechanisms and compliance tools.

Provide architectural guidance and design patterns for leveraging generative AI within critical business processes like License Audit and Product Approval, emphasizing scalable and verifiable outcomes.

Establish and automate DevOps pipelines for model deployment, testing and iterative enhancements.

Skills :

Deep expertise with Azure AI services (including Cognitive Services) and Google Cloud AIML platform.

Extensive knowledge and practical experience with document processing using Azure Document Intelligence and alternative OCR technologies.

Mastery of Snowflake for data modeling, warehousing, and integrating with advanced analytics workflows.

Proficiency with cloud orchestration tools, like GKE Scheduler, and asynchronous communication services (e.g., Google PubSub).

Advanced expertise in API management using Apigee and related security best practices.

Proven ability to design and implement Kubernetes-based AI model deployments, including model optimization techniques.

Strong command of Python for development and automation tasks.

Demonstrated experience converting business needs into scalable and maintainable technical designs.

Experience working with and extending custom Generative AI tooling.

Deep understanding of Retrieval Augmented Generation (RAG) concepts and its applications.