Frontier AI & Agentic Automation Intern (People Technology)

Your Role:
The People Technology team is building the next generation of HR services using Autonomous AI Agents grounded in enterprise knowledge. We are seeking a highly skilled Frontier AI & Agentic Automation Intern to focus on prototyping solutions that integrate and extend leading models and platforms: Google Gemini, Anthropic Claude, Microsoft Copilot Studio, and Glean.

This is a high-impact, strategic role. The intern will directly contribute to how we leverage secure, cross-application knowledge retrieval (via Glean) to inform the reasoning of cutting-edge LLMs (Gemini, Claude) and build custom, action-taking Copilot agents that revolutionize employee experience and HR efficiency.

Key Responsibilities

• Enterprise Grounding and RAG Integration:
• Design and implement Retrieval-Augmented Generation (RAG) pipelines that leverage the Glean platform’s knowledge graph and secure search API to provide highly accurate, permission-aware context to LLMs for HR-related queries.
• Focus on the security and compliance aspects of grounding AI answers in sensitive employee data (e.g., policies, payroll data).
• Multi-Platform Agent Orchestration:
• Prototype and test agents that use different LLM backends (e.g., Gemini API for multimodal analysis, Claude API for complex policy reasoning) based on the specific requirements of the HR task.
• Explore the use of Microsoft Copilot Studio to create custom tools (plugins) and orchestration flows that integrate with existing People Tech systems and can be leveraged within the Microsoft 365 environment.

Tool Development for Agents:

• Build custom, encapsulated actions (tools) using Python or JavaScript to enable Agents to perform functions within HR systems (e.g., submitting a ticket, updating a profile field, retrieving a specific document based on Glean search results).
• Specifically, investigate the Model Context Protocol (MCP) if applicable, to enable seamless interoperability between Glean knowledge and Copilot agents.
• Performance & Ethics Benchmarking:
• Develop a testing methodology to compare the accuracy, helpfulness, and safety of Agent responses when using different LLM providers and grounding sources.
• Document the decision-making logic and failure modes of autonomous agents to ensure auditability and responsible AI practices.

Required Skills & Qualifications

• Current Enrollment: Pursuing a Master’s or advanced Bachelor’s degree in Computer Science, AI/ML, Data Science, or a closely related field.
• Understanding of HR data security, PII (Personally Identifiable Information) handling, and relevant compliance frameworks.
• Excellent verbal and written communication skills with strong time management ability.
• Ability to work independently and collaboratively within a team.
• Strong organizational skills and the ability to manage multiple tasks simultaneously.
• Must be able to work in a fast-paced, dynamic environment, under pressure, and with a high degree of professionalism, delivering high-quality results on time.
• Proficient in MS Office 365 and willing to learn internal systems and other technical tools.
• Programming Proficiency: Strong proficiency in Python for building APIs, agents, and custom tools.

LLM & Agentic Experience:

• Direct experience using the APIs of Gemini, Claude, or other frontier LLMs for structured data generation and complex reasoning.
• Demonstrated understanding of AI Agent architectures (Planner, Tools/Actions, Memory, RAG).

Technical Fundamentals:

• Practical experience working with REST APIs and handling structured data (JSON, YAML).
• Familiarity with at least one enterprise AI tool or framework (e.g., LangChain, or prior exposure to Copilot Studio or Glean).

Desired (But Not Required) Skills

• Hands-on experience building custom Plugins/Tools for Microsoft Copilot Studio or similar platforms (e.g., Power Automate, Power Apps connectors).
• Experience with Glean Search or other sophisticated enterprise knowledge graphing/semantic search platforms.
• Experience with cloud platforms (Azure, GCP, or AWS) and containerization (Docker).

Attractive Package

• Friendly, fast-paced atmosphere
• Collaborative diverse working environment in a growing company
• Work with innovative technologies & solutions

We work in hybrid mode 3+2, a minimum of 3 days at the office with mandatory Tuesdays and Wednesdays and a maximum of 2 days remote per week.