Sr AI Engineer

Job Title: Senior AI Engineer – Google AI & Generative Intelligence Duration: 6 Months (Temp-to-Hire) Location: Paramus, NJ [Hybrid] Role Overview We are seeking a highly experienced Senior AI Engineer with deep expertise in Google AI technologies, Generative AI. The ideal candidate brings 10–15 years of broad software engineering experience, with the last 4+ years focused exclusively on Artificial Generative Intelligence, including designing, building, deploying, and monitoring production-grade AI systems. This role demands mastery of the Google ecosystem — including Google Workspace, Google Agent Development Kit (ADK), and Vertex AI — alongside a strong command of modern LLM/SLM frameworks, cloud-native infrastructure, and MLOps best practices. Key Responsibilities 1. Large & Small Language Model Engineering Design, develop, and deploy Agents leveraging commercial LLMs such as Gemini (Google), GPT (OpenAI), and Claude Sonnet (Anthropic) for high-performance, large-context, and multimodal tasks. Work with open-source/self-hosted LLMs including Mixtral (Mistral AI). Architect and implement SLM-based solutions using lightweight models such as Phi-3 (Microsoft), Gemma (Google), and Mistral for resource-constrained environments. Lead fine-tuning and customization of models using Vertex AI Tuning, Hugging Face Transformers, and parameter-efficient fine-tuning (PEFT) methods including LoRA and QLoRA. Apply training frameworks such as PyTorch, TensorFlow, or JAX for model experimentation and development. Generate synthetic data and evaluate models using HELM, lm-evaluation-harness, and custom benchmarks. 2. Google AI & Workspace Integration Lead the design and implementation of AI-powered solutions deeply integrated with Google Workspace (Docs, Sheets, Drive, Gmail, Meet), Big Query and Lakehouse. Architect and build intelligent agents and workflows using Google Agent Development Kit (ADK). Leverage Google AI Studio as the primary IDE, VSCode for AI application development and prototyping. Utilize Google Cloud Platform (GCP) services including: Vertex AI for ML model training, tuning, and deployment GKE (Google Kubernetes Engine) for container orchestration Cloud Run for serverless deployment Cloud Functions for event-driven AI tasks Vertex AI Vector DBs for semantic search and retrieval 3. Design & Planning Lead requirements gathering using Confluence for documentation and team collaboration. Create detailed system architecture diagrams and AI workflows using Lucidchart. Design UI/UX prototypes in Figma for AI-powered application interfaces. Manage project delivery and sprint planning using Jira. Oversee data preparation and management: cleaning, transforming, and organizing data for AI/ML workflows. Conduct data analysis using Jupyter Notebooks and pandas for exploration and preprocessing. Leverage Hugging Face Model Hub for model comparison, selection, and download. 4. Development Frameworks & Tools Orchestrate LLM/SLM applications using LangChain, LlamaIndex, and LangGraph. Build multi-agent systems with Semantic Kernel, and LangGraph. Manage and optimize prompts using LangSmith and PromptLayer. Deploy models locally with Ollama or at scale with vLLM for efficient inference. Track experiments, metrics, and results with MLflow or Weights & Biases. Manage code and data versioning with Git. 5. Vector Databases & Semantic Search Implement semantic search and Retrieval-Augmented Generation (RAG) pipelines using Vertex AI Vector DBs and ChromaDB. Design and optimize end-to-end RAG architectures for enterprise-grade knowledge retrieval. 6. Backend Development Develop robust RESTful APIs using FastAPI (Python) or Express.js (Node.js). Manage and secure APIs using Mulesoft, Apigee. 7. Frontend Development Build modern user interfaces using React or Angular. Utilize Material-UI for consistent, accessible, and modern UI components. Prototype and plan UI/UX workflows using Figma. 8. Development Tools & Code Quality Write and debug code in VS Code with Python and GitHub Copilot extensions. Leverage GitHub Copilot for AI-assisted code suggestions and productivity. Manage source code with GitHub or GitLab. Enforce code quality and standards using SonarQube, ESLint, and Pylint. 9. Testing & Quality Assurance Conduct LLM-specific testing using RAGAS and DeepEval for LLM/RAG pipeline evaluation. Use LangSmith Evaluators for prompt testing and hallucination detection. Write and execute unit tests using pytest. Ensure output quality and reliability using LangChain Evaluators and custom metrics. 10. Deployment & Infrastructure Orchestrate containers at scale with Kubernetes (K8s), and Google GKE. Automate CI/CD pipelines using GitHub Actions or GitLab CI. Support on-premise, cloud (GCP/Vertex AI), and hybrid infrastructure deployments including edge devices for local inference. 11. LLM Monitoring & Observability Monitor LLM performance and usage with LangSmith and Weights & Biases. Track and optimize AI in