Software Engineer (AI Agentic Solutions), Tech Consulting

EY is recruiting a Software Engineer (AI Agentic Solutions) within Technology Consulting in Ebene, Mauritius.

The opportunity

We are looking for a hands-on developer with strong expertise in Golang and Python who can design and build AI-powered solutions and autonomous agents end to end. You will work on integrating LLMs into production systems — APIs, data pipelines, and agentic workflows — and will be expected to leverage modern AI coding tools (Claude Code, GitHub Copilot, Cursor) as a force multiplier in your daily development.

Key Responsibilities

• Design, develop, and deploy applications and microservices in Golang and Python that leverage AI/ML capabilities
• Architect and build AI agents — multi-step reasoning workflows, tool/function calling, memory, planning, and orchestration — using frameworks like LangChain, LangGraph, CrewAI, or custom agentic loops
• Build and integrate solutions using LLM APIs (Anthropic, OpenAI, etc.) including RAG pipelines, prompt engineering, and vector database integrations (pgvector, Pinecone, FAISS)
• Use AI-assisted development tools (Claude Code, GitHub Copilot, Cursor) effectively for rapid prototyping, refactoring, test generation, and agent-driven coding workflows
• Develop high-performance backend services and APIs (Gin/Echo/gRPC in Go; FastAPI in Python)
• Write clean, testable, production-grade code with proper version control (Git) and CI/CD practices
• Collaborate with cross-functional teams to translate business problems into AI-driven solutions

To qualify, candidates must have

• A related bachelor's degree in Computer Science, Information Technology, or a related field or any STEM degree
• Approximately 3-6 years of experience in a related field will be an advantage

Ideally, you will also have

• Strong hands-on Golang development: goroutines, channels, concurrency patterns, building high-performance microservices, gRPC, and CLIs
• Strong hands-on Python development (3+ years): OOP, async programming, common libraries (Pandas, NumPy)
• Deep practical understanding of building AI agents — agent architectures (ReAct, plan-and-execute), tool calling, MCP (Model Context Protocol), context/memory management, and agent evaluation
• Demonstrated experience using AI coding tools as part of the development workflow
• Solid AI/ML fundamentals: how LLMs work, embeddings, fine-tuning vs. RAG, evaluation basics
• Working knowledge of SQL/NoSQL databases; exposure to vector databases is a plus
• Familiarity with cloud platforms (AWS/Azure/GCP), Docker, and Kubernetes basics