Accenture

GenAI Full-Stack Engineer/ Agentic App Engineer

Accenture
ConsultingSingaporeOnsitePosted 2 weeks ago

About the role

AI summarised

Design and build end-to-end, AI-powered applications with a focus on rapid MVPs and production-ready agentic experiences. This role involves working across the entire stack—from LLM integration and RAG pipelines to APIs and front-end UIs—to deliver high-impact generative AI solutions for clients.

ConsultingOnsiteOther Functions

Key Responsibilities

  • Build full-stack MVP and production-quality applications utilizing agentic AI patterns and multi-agent frameworks.
  • Design and implement LLM-based features, including prompt engineering, tool usage, and Retrieval-Augmented Generation (RAG) pipelines.
  • Develop backend services and APIs using Python and FastAPI, integrating with vector databases and external tools.
  • Create intuitive front-end user interfaces using React or Next.js to deliver compelling agentic experiences and demos.
  • Implement and configure multi-agent workflows using LangGraph and related frameworks, including routing, tool integration, and coordination.
  • Build reusable components, libraries, and templates to accelerate future GenAI and agentic builds.
  • Collaborate closely with product managers, designers, and AI engineers to deliver end-to-end features from concept to deployment.

Requirements

  • Strong experience in full-stack development using Python for backend and React/Next.js for frontend.
  • Hands-on experience with LLM integration, prompt engineering, and RAG pipeline implementation.
  • Familiarity with agent frameworks and orchestration tools such as LangChain, LangGraph, or similar multi-agent systems.
  • Experience building APIs and microservices with FastAPI or equivalent frameworks.
  • Knowledge of containerization and modern delivery practices (Docker, Git, CI/CD).
  • Experience working with vector databases and integrating them into AI applications.
  • Strong debugging skills and a track record of rapid prototyping for MVPs and demos.