| Mandatory skill - Agentic AI Frameworks Langchain, LangGraph, CrewAI, Vertex.AI, Autogen etc.) |
| Good to have skills can be mentioned here : Cloud Native, Microservice Architecture Role: Gen AI Architect New Jersey Full Time Only - 15 years in software/solution architecture, proven experience as a Data Scientist or ML Engineer with exposure to agent-based AI systems.
- Design & Implement Agentic AI systems using agent frameworks (AutoGen, LangGraph, CrewAI, etc.) to build multi-agent and goal-oriented systems.
- Proficiency in Prompt Engineering, few-shot prompting, chain-of-thought reasoning, and prompt templates.
- Familiarity with cloud-native AI platforms from AWS, Azure, or Google Cloud Platform (e.g., Bedrock, Azure OpenAI, Vertex AI).
- Experience on AI for Engineering & working with AI Code Assist tools (e.g., Copilot, Windsurf, Cursor)
- Experience working with Vector databases & design & deploy RAG pipelines, MCP Servers & A2A Implement robust LLMOPs - for continuous integration, deployment, monitoring, logging, and troubleshooting mechanisms for GenAI applications.
- Develop and promote reusable architectural patterns, best practices, and governance frameworks for GenAI development.
- Lead the end-to-end architectural design of Generative AI applications, ensuring scalability, performance, security, cost-effectiveness, and maintainability.
- Proficiency in containerization technologies (Docker, Kubernetes) and CI/CD pipelines.
- Must have experience working with Microservice architecture using Spring Boot Rest APIs & know API Security, Versioning
- Must have experience designing CloudNative applications on any cloud such as AWS, Azure, Google Cloud Platform, Spring Cloud, PCF
- Programming proficiency in Python (preferred), and optionally Java/Node.js for integration.
- Collaborate with Data Scientists, Product Owners, and Business SMEs to translate business problems into AI-powered solutions.
- Exceptional communication and presentation skills, with the ability to articulate complex technical concepts to both technical and non-technical audiences
|
