Senior Staff Engineer, Generative Ai at Nagarro1

REQUIREMENTS: • Total experience 8+ years. • Deep understanding of LLMs (e.g., GPTs, Llama, Claude, Gemini, Qwen, Mistral, BERT-family models) and their architectures (Transformers) • Should have expert-level prompt engineering skills and proven experience implementing RAG patterns • High proficiency in Python and standard AI/ML libraries (e.g., LangChain, LlamaIndex, LangGraph, LangSmith, Hugging Face Transformers, Scikit-learn, PyTorch/TensorFlow). • Experience implementing RAG architectures and prompt engineering. • Strong experience with fine-tuning and distillation techniques and evaluation. • Strong experience using managed AI/ML services on the target cloud platform (e.g., Azure Machine Learning Studio, AI Foundry). • Strong understanding of vector databases (e.g., Weaviate, Neo4j) • understanding of GenAI evaluation metrics (e.g., BLEU, ROUGE, perplexity, semantic similarity, human evaluation). • Architect and implement scalable GenAI and Agentic AI solutions end-to-end. • Should be able to write high-quality, production-ready Python code with strong testing and maintainability practices. • Should be able to productionize AI systems on Azure or AWS, ensuring enterprise-grade reliability and performance. • Should be able to build and expose APIs using FastAPI, integrating with databases through an ORM. • Should be able to scale GenAI solutions to support enterprise workloads. • Collaborate across product and engineering teams to convert business needs into AI-driven solutions. • Strong ability to both architect and code GenAI/Agentic AI solutions. • Proven production experience with GenAI deployments on Azure or AWS. • Should be able to build & deploy AI pipelines using SageMaker, Vertex AI, or Azure ML • Hands on Docker, Kubernetes, and CI/CD pipelines (GitHub Actions, Argo) for scalable AI infra • Hands-on with serverless AI APIs, containerized model serving, and GPU orchestration • Experience with IaC (Terraform / Bicep) and cloud monitoring tools • Data pipelines via Airflow, Kafka, or Databricks • Strong experience in scaling AI solutions in live environments. • Very strong Python programming skills with a track record of clean, efficient, and maintainable code. • Should have successfully delivered at least one production GenAI/Agentic AI solution. • Must have proficiency with FastAPI and at least one ORM (e.g., SQLAlchemy, Tortoise ORM). • Should have experience with Model Context Protocol (MCP). • Should have contributions to open-source GenAI projects. • Good to have experience with React (or some other JS frameworks) for building user-facing interfaces and front-end integrations • Excellent communication skills and the ability to collaborate effectively with cross-functional teamsRESPONSIBILITIES: • Understanding the client’s business use cases and technical requirements and be able to convert them into technical design which elegantly meets the requirements. • Mapping decisions with requirements and be able to translate the same to developers. • Identifying different solutions and being able to narrow down the best option that meets the clients’ requirements. • Defining guidelines and benchmarks for NFR considerations during project implementation. • Writing and reviewing design document explaining overall architecture, framework, and high-level design of the application for the developers. • Reviewing architecture and design on various aspects like extensibility, scalability, security, design patterns, user experience, NFRs, etc., and ensure that all relevant best practices are followed. • Developing and designing the overall solution for defined functional and non-functional requirements; and defining technologies, patterns, and frameworks to materialize it. • Understanding and relating technology integration scenarios and applying these learnings in projects. • Resolving issues that are raised during code/review, through exhaustive systematic analysis of the root cause, and being able to justify the decision taken. • Carrying out POCs to make sure that suggested design/technologies meet the requirements.