Job Purpose

This internship is focused on the design, development, and evaluation of large language model (LLM) systems and agentic AI applications in applied energy and engineering contexts. The intern will contribute to active projects centered on intelligent workflow automation, including agentic systems for commissioning, field service, sales support, and engineering assistance. Depending on project needs and the intern's background, engagement with adjacent areas — including optimization, simulation, and deep learning-based forecasting — is also expected.

Essential Duties and Responsibilities

• Design and implement agentic AI architectures, including multi-agent workflows, tool-calling systems, memory and state management, and orchestration logic for complex multi-step tasks.
• Develop and evaluate retrieval-augmented generation (RAG) pipelines, with attention to retrieval strategy, document chunking and indexing, embedding model selection, re-ranking, and end-to-end evaluation.
• Contribute to LLM evaluation and validation frameworks — defining test coverage, constructing evaluation datasets, assessing output reliability, and identifying failure modes through structured testing and adversarial analysis.
• Conduct prompt engineering and instruction design for domain-specific tasks, and support experimentation with parameter-efficient fine-tuning approaches where applicable.
• Perform model benchmarking and comparative analysis, including evaluation of commercial and open-source LLMs for specific task types, latency and cost tradeoffs, and domain adaptation requirements.
• Support integration of LLM and agentic components with broader system architectures, including data pipelines, APIs, and domain-specific tooling.
• Contribute to data preparation and preprocessing workflows for structured and unstructured industrial datasets, including cleaning, transformation, and schema design.
• Engage in the full engineering rigor expected of production AI/ML systems — including unit and integration testing, model verification and validation, experiment evaluation, simulation workflows, data and output visualization, and technical documentation and reporting — as continuous activities throughout the project lifecycle.

Qualifications

• Currently enrolled in a Master's or PhD program in Computer Science, Artificial Intelligence, Data Science, Electrical Engineering, or a closely related discipline — or recently graduated from such a program.
• Strong theoretical foundation in machine learning and deep learning, with the ability to reason about model behavior, generalization, and failure modes.
• Demonstrated hands-on experience building LLM-based applications — including at least one of the following: agentic systems, RAG pipelines, structured output generation, or tool-augmented language models.
• Proficiency in Python; fluency with ML frameworks, particularly PyTorch.
• Strong data handling and analysis skills — experience working with complex, real-world datasets using pandas, NumPy, or equivalent tools.
• Ability to design and execute rigorous experiments, interpret results critically, and communicate findings clearly in written and verbal form.

Preferred Qualifications

• Experience building LLM-based applications, including agentic workflows, tool-augmented models, or retrieval-augmented generation systems; familiarity with agent frameworks (such as LangChain, CrewAI, or similar) is a plus.
• Exposure to optimization or simulation methods (e.g., linear programming, MPC, or heuristics), particularly in energy, industrial, or cyber-physical system contexts.
• Experience with time-series data or sequential modeling tasks, including forecasting, anomaly detection, or sequence classification.
• Familiarity with cloud-based AI/ML platforms, particularly Microsoft Azure (e.g., Azure ML or Azure AI Foundry); experience deploying or evaluating LLMs in a cloud environment is a plus.
• Interest in or prior exposure to energy storage systems, battery management, industrial automation, or related domains is a plus.

General Job Requirements