GenAI SME (CoPilot / ChatGPT)

About the Role

Were looking for a seasoned GenAI SME to lead solution discovery, design, and implementation of Generative AI capabilities across business workflows. You will act as the trusted advisor for stakeholders, drive production-grade deployments, and establish governance and performance baselines to deliver measurable business impact.

This role sits at the intersection of AI engineering, product, and business consultingideal for someone who can translate business problems into GenAI solutions, and take them from prototype pilot production.

Key Responsibilities

Strategy & Advisory

• Partner with business leaders to identify high-value GenAI use cases (e.g., knowledge retrieval, summarization, agentic automation, copilots).
• Perform feasibility assessments, define ROI, TCO, and risk; develop roadmaps with stage gates (POC Pilot Scale).
• Create reference architectures and patterns for RAG, agents, prompt orchestration, and evaluation frameworks.

Solution Design & Delivery

• Lead end-to-end solution delivery using LLM stacks (Azure OpenAI / OpenAI, Google Vertex, Anthropic, etc.).
• Architect RAG pipelines (chunking, embeddings, vector stores, retrieval optimization, reranking).
• Implement prompt engineering and prompt orchestration (templates, tools, memory, guardrails).
• Build agent workflows (task planning, tool-use, function calling, multi-agent collaboration).
• Productionalize with CI/CD, observability (latency, token usage, hit-rate), and cost controls.

Data & Governance

• Collaborate with data teams to prepare corpora and embeddings, metadata strategies, and quality standards.
• Implement evaluation & metrics: relevance, faithfulness, toxicity, bias, hallucination rate, task success.
• Ensure responsible AI compliance: privacy, security, IP, safety, provenance, and audit readiness.

Platform & Enablement

• Contribute to or manage internal GenAI platform (APIs, templates, guardrails, evaluation harness).
• Create playbooks, reusable components, and documentation; conduct enablement workshops for stakeholders.
• Mentor L2/L1 engineers and analysts; review solutions and ensure best practices.

Required Qualifications (L3)

• 712 years total experience, with 3+ years hands-on in GenAI/LLMs (production deployments preferred).
• Language skills:
• French: Fluent
• English: Intermediate
• Strong in at least two of:
• RAG (document processing, embeddings, vector DBs like Azure AI Search, Pinecone, Weaviate, Qdrant).
• Agentic frameworks (LangChain, LlamaIndex, Semantic Kernel, DSPy).
• Evaluation & Safety (automatic evals, structured outputs, guardrails, content filters).
• Hands-on with Azure OpenAI (preferred) and one additional LLM provider (OpenAI, Anthropic, Google).
• Proficiency in Python; experience building APIs/microservices (FastAPI/Flask), CI/CD, and observability.
• Experience deploying on Azure (Functions, App Service, AKS, Storage, Key Vault, Monitor), or equivalent cloud.
• Strong prompt engineering and orchestration skills (function calling, tools, templates, system prompts).
• Familiarity with document ingestion pipelines (PDF/HTML/Office), chunking strategies, and reranking.
• Solid understanding of security & compliance: PII handling, data residency, RBAC, private networking, secrets.
• Excellent stakeholder communication; able to translate business requirements into technical designs.

Preferred/Good-to-Have

• Experience with M365 Copilot/Graph connectors, SharePoint indexing, Teams integrations.
• Knowledge of vector DB ops (indexing strategies, HNSW/IVF, recall vs. precision tuning).
• Background in ML Ops or Data Engineering for AI products.
• Exposure to evaluation datasets creation and human-in-the-loop review processes.
• Basic knowledge of genAI cost modeling and optimization (caching, prompt compression, distillation).