[Intern] Agentic AI for Conversational Data Analytics

You will join our data science team in developing agentic AI and conversational analytics capabilities for policy analysis. These capabilities allow users to query and analyse structured data, survey data and housing market news using natural language, and include AI assistants for statistical modelling and forecasting. Several related projects may run concurrently. Interns will be assigned to one or more projects based on project needs and their skills. The solutions will be developed primarily in Python on AWS SageMaker, leveraging cloud-based data, analytics and AI services. Responsibilities Depending on the assigned project, responsibilities may include: - Developing a coordinating layer that interprets users’ questions, selects appropriate data sources, retrieval methods and analytical tools, and combines their outputs into a coherent response. - Converting existing analytics and modelling capabilities into reusable tools, and exploring the use of Model Context Protocol (MCP) to connect them through a common interface. - Improving the housing market news and sentiment analytics application, including its data processing, retrieval and analytical capabilities. - Exploring more autonomous AI agents that can plan, execute and validate analyses with human oversight. - Developing an offline conversational analytics solution using open-weight language models. These projects build on RSM’s existing conversational data-analysis prototypes and planned integration, orchestration and autonomous-analysis workstreams. Requirements - Good foundation in Python, SQL, data processing and statistical analysis. - Ability to use AI coding tools effectively, while applying sound judgement to understand, test and validate their outputs. - Ability to quickly digest technical documentation and research on large language models, agentic workflows, information retrieval and conversational analytics, and critically assess, adapt and integrate relevant approaches into working solutions. - Ability to work independently and collaborate with the project team. Special Considerations: Third- or fourth-year university students studying Computer Science or Data Science. Strong second-year students with relevant technical experience may also be considered. Applicants who are available only on a part-time basis during the academic term must be able to continue the internship on a full-time basis during the subsequent summer vacation. Experience with AWS SageMaker, APIs, agent frameworks, information retrieval, open-weight language models or statistical modelling would be useful, but is not essential. Candidates should submit both their CV and latest academic transcript as part of their application.