GenAI Fullstack Engineer

Job Description

Marlabs, a global AI and Digital Solutions Consulting firm, delivers intelligent solutions across AI, data, analytics, and product engineering. Since 2000, we have partnered with some of the largest healthcare, life sciences, financial services, and government organizations worldwide. As we continue to expand our global footprint, we have an exciting opportunity for a highly skilled GenAI Fullstack Engineer to join our innovative and dynamic team.

Employment Type : Full-time

Location : Onsite- Englewood Cliffs, NJ

GenAI Fullstack Engineer | About the Role

As a GenAI Fullstack Engineer, you will design and build scalable enterprise applications that combine robust Java-based backend engineering with modern Generative AI capabilities. You will work across Java, Spring Boot, Microservices, Python, FastAPI, LLMs, RAG, and Vector Databases to develop production-ready AI services and integrate them with enterprise applications. The ideal candidate is a strong software engineer who has expanded into hands-on GenAI engineering and can develop secure, scalable solutions spanning traditional backend services, AI orchestration, enterprise knowledge retrieval, and LLM-powered applications.

GenAI Fullstack Engineer | Day-to-Day

\n
• Design, develop, and maintain scalable backend applications and REST APIs using Java, Spring Boot, Microservices, Spring Security, and Spring Data/JPA.
\n
• Build and integrate production-grade GenAI and LLM applications, connecting AI services with existing Java/Spring Boot enterprise applications and business systems.
\n
• Design and implement Retrieval-Augmented Generation (RAG) pipelines, including document ingestion, chunking, embeddings, indexing, semantic search, context retrieval, reranking, and response generation.
\n
• Develop AI services and orchestration components using Python and FastAPI, leveraging frameworks such as LangChain, LangGraph, LlamaIndex, or Hugging Face to support reusable enterprise AI capabilities.
\n
• Implement and optimize Vector Databases such as Pinecone, Weaviate, OpenSearch, FAISS, or ChromaDB to enable enterprise knowledge retrieval and semantic search.
\n
• Integrate LLM platforms such as Azure OpenAI, OpenAI, AWS Bedrock, Claude, Gemini, or Llama, applying prompt engineering, response evaluation, guardrails, and optimization techniques to improve solution accuracy and reliability.
\n
• Build secure and observable AI/backend services by implementing authentication, authorization, exception handling, logging, monitoring, API security, and production-grade engineering standards.
\n
• Containerize and deploy applications using Docker and CI/CD pipelines, collaborating with architects, data engineers, developers, and business teams to deliver reliable and scalable GenAI solutions.
\n

GenAI Fullstack Engineer | Skills & Experience

\n
• Strong software engineering 3+ years experience with Java, Spring Boot, Microservices, and REST APIs, including hands-on experience designing scalable and distributed enterprise backend applications.
\n
• Hands-on experience developing production GenAI/LLM applications, including LLM integration, prompt engineering, response evaluation, and enterprise AI use cases.
\n
• Strong production experience with RAG architectures, including embeddings, document ingestion, chunking, semantic retrieval, reranking, context construction, and RAG quality optimization.
\n
• Strong working knowledge of Python and FastAPI, with experience building AI/ML services and integrating Python-based AI components with Java/Spring Boot applications.
\n
• Hands-on experience with Vector Databases and GenAI frameworks, such as Pinecone, Weaviate, OpenSearch, FAISS, ChromaDB, LangChain, LangGraph, LlamaIndex, or Hugging Face.
\n
• Experience with cloud and modern deployment practices, including AWS, Azure, or GCP, Docker, CI/CD, monitoring, observability, and production deployment of AI services.
\n
• Additional exposure to Agentic AI, MCP (Model Context Protocol), AI guardrails, LLMOps, Kubernetes, knowledge graphs, multimodal AI, and event-driven microservices is advantageous.
\n