Internship / Master Thesis : Going beyond conversational chatbots at ELCA Switzerland and Europe

Pully, Vaud, Switzerland -

Full Time

Start Date

Immediate

Expiry Date

08 Jun, 26

Salary

0.0

Posted On

10 Mar, 26

Experience

0 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Large Language Models, Generative AI, Python, Web Technologies, Agentic Architectures, LLM-based Systems, UI/Agent Contracts, Prototyping

Industry

IT Services and IT Consulting

Description

With the rise of Generative AI and LLM, we have seen a lot of chatbot-oriented user interface like Gemini, ChatGPT or Perplexity, we have been able to build our own chatbot oriented UIs with solutions like OpenWebUi or LibreChat. But actually, textual or even multimodal conversations are likely not the most productive and efficient way for business users to interact with business applications. This internship aims at exploring how traditional web human user interfaces (involving forms, grids, panels, buttons etc…) can be built on top of an agentic server-side business layer and combine the productivity of traditional UIs with power of AI. During this internship, you will conduct applied research on how agentic, LLM-based systems can support structured enterprise user interfaces beyond conversational chat. You will: • Formulate research questions and hypotheses on the limitations of conversational UIs for business workflows and the potential benefits of agent-backed structured interfaces. • Study and compare agentic architectures (planner–executor, tool-calling, stateful agents) and their suitability for driving deterministic, auditable business logic. • Propose and formalize a novel interaction model between an agentic business layer and traditional web UI components, including explicit representations of intent, state, and available actions. • Design and evaluate UI–agent contracts (schemas, protocols, metadata) enabling dynamic generation or adaptation of forms, tables, and workflows. • Implement experimental prototypes applying these models to representative business processes and iteratively refine them based on observed limitations. • Define and apply evaluation criteria (task completion, error rates, predictability, cognitive load) to compare chat-based and structured agent-driven interfaces. • Analyze results to identify design patterns, trade-offs, and architectural principles for enterprise-grade agent-backed UIs. • Produce a report synthesizing the research contributions, experimental results, and perspectives for future work. Our offer • A dynamic work and collaborative environment with a highly motivated multi-cultural and international sites team • The chance to work on what could be the next generation of architecture. • Various internal coding events (Hackathon, Brownbags), see our technical blog • Monthly After-Works organized per location • Good work-life balance (2 days per week from home) Skills required • Interest in Large Language Models (GPT, Claude…) and Generative AI in general • Some proficiency with Python and web technologies

Responsibilities

The intern will conduct applied research on integrating agentic, LLM-based systems with traditional structured enterprise user interfaces, moving beyond conversational chatbots. Key tasks involve formulating research questions, studying agentic architectures, proposing novel interaction models, and designing UI-agent contracts.