Ai Ml Engineer (Core Machine Learning Expertise, Python, Generative A

**1.** **Generative AI (LLMs & RAG)** - 35+ years handson with **LLMs** (Azure OpenAI, OpenAI, Anthropic, Llama family, Cohere or equivalent) - Designing **RAG pipelines** with embeddings, vector stores (e.g., **Azure AI Search**, FAISS, Pinecone, Weaviate, chromadb,qdrant), chunking strategies, citation grounding, and promptsafe retrieval - **Prompt engineering** and **prompt lifecycle** (versioning, testing, safety, guardrails) - **Evaluation & benchmarking** of LLM quality (faithfulness, relevance, toxicity, jailbreak resilience); A/B testing & regression baselining **2. MCP Model Context Protocol (Tools & Agents)** - Building **MCP servers/clients** to expose tools (APIs, DBs, file systems, search, graph) to LLMs - Designing **tool contracts** (schema, auth, rate limits, idempotency), resource providers, and **secure execution sandboxes** - Orchestrating **multitool workflows** via MCP with tracing, retries, and timeout strategies - Experience integrating MCP with **agent frameworks** (e.g., LangChain/ LangGraph, Llama Index, Google Agent ADK) - **Architect & deliver** agentic GenAI applications leveraging **MCP tools** and **FastAPI** microservices for enterprise use cases - Design **secure tool ecosystems** (MCP servers) exposing internal capabilities to LLMs with leastprivilege access, quotas, and auditability - Build **RAG systems** with robust grounding, hallucination control, and **retrieval quality** monitoring - Establish **GenAIOps**: prompt/model/dataset versioning, canary releases, offline/online evaluations, telemetry & feedback loops - Implement **cost, latency, and reliability SLOs**; optimize model choices, caching, batch inference, and streaming UX - Lead **code/design reviews**, mentor engineers, and drive standards for **observability, testing (unit/e2e/redteam), and security** - Partner with product, architecture, and platform teams to translate requirements into **scalable, compliant** solutions - Produce **design docs, runbooks, and threat models**; champion **Responsible AI** practices across the lifecycle Note: This is a third party job (Aggregated by careeruplift.ai). Shortlisting and Final hiring decision & process is handled by the company.