Source description
About the role
Consultant Software Engineer / AI Application Developer Type: Full-Time | Level: Mid to Senior (7–10 Years of Experience) Key Responsibilities
- AI Application Design & Development • Design and develop AI-powered applications using LLMs (OpenAI, Anthropic, Azure OpenAI, etc.) • Build and maintain RAG (Retrieval-Augmented Generation) pipelines including vector stores, chunking strategies, and embedding models
- Agentic Solutions & Orchestration • Develop agentic solutions using frameworks such as LangChain, LangGraph, AutoGen, CrewAI, or Semantic Kernel • Implement and integrate MCP (Model Context Protocol) servers and clients for tool-enabled LLM workflows • Build and expose tools, APIs, and data connectors for use within agent runtimes • Work with orchestration patterns: multi-agent systems, planning loops, memory management, and tool-use
- Quality, Evaluation & Prompt Engineering • Evaluate LLM output quality, implement guardrails, and optimize prompt engineering
- Full Stack Development • Design and develop full stack applications spanning backend (Python / .NET) and frontend (React) layers
- Cloud & Production Deployment • Architect and deploy solutions on cloud platforms (AWS and/or Azure) • Collaborate with data engineers and backend teams to integrate AI capabilities into production systems
Required Skills Core Experience • 7–10 years of overall software development experience • 1–3 years of hands-on experience working in the AI/ML domain • Strong full stack development experience: Python and/or .NET on the backend, React on the frontend Technical Expertise • Cloud platform experience: AWS (Bedrock, Lambda, S3, etc.) and/or Azure (Azure OpenAI, Azure ML, AKS, etc.) • Hands-on experience building RAG pipelines (LlamaIndex, LangChain, or custom) • Experience with vector databases (Chroma, Qdrant, Weaviate, pgvector, etc.) • Familiarity with MCP and tool/function calling patterns in LLM applications • Experience designing agentic workflows (ReAct, plan-and-execute, multi-agent) • REST API development (FastAPI / Flask / ASP.NET) • Understanding of prompt engineering, context window management, and LLM evaluation
Nice to Have • Familiarity with observability tools for LLM apps (LangSmith, Arize, Helicone) • Knowledge of fine-tuning or RLHF workflows • Exposure to MCP server development (SSE / stdio transport) • Containerization with Docker / Kubernetes • Experience working in the cyber domain or having cyber knowledge will be an added advantage
More at Societe Generale