Principal Data Scientist - R01560286

Primary Skills

• A/B Testing, Clustering, AWS, Hypothesis Testing, T-Test, Z-Test, Regression (Linear, Logistic), Parameter Tuning, Data Wrangling, Exploratory Data Analysis, Python/PySpark, PCA, Factor Analysis, SAS/SPSS, GitHub / Gitlab, IBM Watson, Feature Engineering, Tree Aglorithms, Grid Search, SVM, Tools(KubeFlow, BentoML), Cross Validation, Data Curiosity, Design Thinking, Data Literacy

Job requirements

• This role is explicitly focused on building agentic systems from scratch, including the design and development of autonomous AI agents and enterprise copilots, with hands-on experience in Microsoft 365 Copilot and Copilot Studio ecosystems as a core requirement. ________________________________________ Core Responsibilities • Agentic AI Architecture & Engineering: Build AI agents and multi-agent systems from scratch, including agent orchestration, autonomy design, tool-use, memory management, and decision frameworks. • Enterprise Copilot Development: Design, build, and scale enterprise copilots using Microsoft Copilot Studio and Microsoft 365 Copilot, enabling intelligent task automation and knowledge workflows. • Build AI agents & copilots on Microsoft Copilot Studio • Develop automated workflows using Power Automate / Logic Apps • Integrate AI agents with enterprise systems via APIs, connectors, and Azure services • Implement RAG, grounding, semantic search using Azure OpenAI & Azure Cognitive Search • Ensure security, governance, and responsible AI practices • Collaborate with architects, analysts, and business teams • Problem Formulation: Translate business objectives into well-defined data science, ML, and Agentic AI problems; validate OKRs using robust statistical and experimental measures. • Agentic AI & LLM Solutions: Design, build, deploy, and optimize Agentic AI systems (multi-agent workflows, task orchestration, autonomous decision-making) using LLMs for real-world enterprise use cases. • LLM Development & Deployment: Fine-tune, prompt-engineer, evaluate, and productionize LLMs (open-source or proprietary) for use cases such as copilots, RAG pipelines, conversational AI, and intelligent automation. • Data Wrangling & Feature Engineering: Handle structured and unstructured data at scale, including text, documents, and conversational data for LLM-powered solutions. • Insight Generation & Data Storytelling: Convert complex analytical outputs and AI model results into clear, compelling narratives for business and executive audiences. • Technical Decision-Making: Make informed trade-offs on model complexity, iteration depth, experimentation cycles, and time-to-value. • Design Thinking & Innovation: Apply design thinking principles to build user-centric AI products and data solutions. • Mentorship & Leadership: Coach senior data scientists, review architectures, and establish best practices across data science, ML, and GenAI initiatives.