**Job Description:**
**Role Overview:**
As a Senior / Lead Data Scientist specializing in Generative AI (RAG & Multi-Agent Systems) at Material, your primary responsibility will be to build and scale an enterprise-grade GenAI chatbot platform. This platform will integrate with multiple RAG systems to provide accurate, grounded, and secure responses at high concurrency. You will be in charge of the architecture and hands-on development across various components like LLM orchestration, retrieval pipelines, evaluation, and production deployment. **Key Responsibilities:**
Design and construct a scalable GenAI chatbot that orchestrates and calls multiple RAG systems for routing, retrieval, re-ranking, and response generation. - Implement multi-agent architectures using frameworks like LangChain, AutoGen, or similar tools. - Develop robust Python services/APIs (preferably using FastAPI) with features like streaming responses, retries, rate limits, and observability. - Build and optimize RAG pipelines including data ingestion, chunking strategies, embeddings, vector indexing, hybrid retrieval, metadata filtering, and context management. - Evaluate and enhance response quality through offline/online evaluation, feedback loops, and guardrails to minimize hallucinations. - Integrate Azure OpenAI / OpenAI GPT models and manage prompt/versioning strategies for different use-cases. - Ensure production readiness by focusing on performance optimization, cost control, caching, monitoring, alerting, CI/CD, and secure deployments. - Collaborate with product and engineering teams to define requirements, SLAs, and rollout plans. **Qualification Required:**
Strong proficiency in Python and AI/ML libraries such as PyTorch, TensorFlow, and Scikit-learn. - Hands-on experience with Generative AI and LLMs like Azure OpenAI GPT in production environments. - Solid understanding of Data Science fundamentals including preprocessing, feature engineering, and model evaluation. - Previous experience in building RAG systems covering aspects like embeddings, vector search, retrieval strategies, re-ranking, and grounding. - Familiarity with orchestration frameworks like LangChain, AutoGen, or similar tools. - Proficient in software engineering practices like clean architecture, testing, code review, and documentation. **Additional Company Details:**
Material is a global strategy partner known for turning customer challenges into growth opportunities. They collaborate with renowned brands and innovative companies worldwide to design and deliver rewarding customer experiences. Material focuses on deep human insights, design innovation, and data to create modern technology-powered experiences that accelerate engagement and growth for their clients. The company is recognized for its expertise in solving complex technology problems and leveraging strategic partnerships with top-tier technology firms. *Note: Additional company details have been provided for better understanding of Material's core values and working environment, enhancing the context of the job role.* **Job Description:**
**Role Overview:**
As a Senior / Lead Data Scientist specializing in Generative AI (RAG & Multi-Agent Systems) at Material, your primary responsibility will be to build and scale an enterprise-grade GenAI chatbot platform. This platform will integrate with multiple RAG systems to provide accurate, grounded, and secure responses at high concurrency. You will be in charge of the architecture and hands-on development across various components like LLM orchestration, retrieval pipelines, evaluation, and production deployment. **Key Responsibilities:**
Design and construct a scalable GenAI chatbot that orchestrates and calls multiple RAG systems for routing, retrieval, re-ranking, and response generation. - Implement multi-agent architectures using frameworks like LangChain, AutoGen, or similar tools. - Develop robust Python services/APIs (preferably using FastAPI) with features like streaming responses, retries, rate limits, and observability. - Build and optimize RAG pipelines including data ingestion, chunking strategies, embeddings, vector indexing, hybrid retrieval, metadata filtering, and context management. - Evaluate and enhance response quality through offline/online evaluation, feedback loops, and guardrails to minimize hallucinations. - Integrate Azure OpenAI / OpenAI GPT models and manage prompt/versioning strategies for different use-cases. - Ensure production readiness by focusing on performance optimization, cost control, caching, monitoring, alerting, CI/CD, and secure deployments. - Collaborate with product and engineering teams to define requirements, SLAs, and rollout plans. **Qualification Required:**
Strong proficiency in Python and AI/ML libraries such as PyTorch, TensorFlow, and Scikit-learn. - Hands-on experience with Generative AI and LLMs like Azure OpenAI GPT in production environments. - Solid understanding of Data Science fun