Internship / Master Thesis : Actionable Interfaces: Bridging Conversational AI and Task Execution

Are you fascinated by the capabilities of Large Language Models (LLMs) like GPT-4 but find yourself wanting more from these intelligent systems? At ELCA, we identify this gap not as a limitation but as an intriguing opportunity. The internship aims to create an architecture where LLMs serve as interactive interfaces for applications. This architecture will be designed for seamless integration and querying of diverse data sources, from structured databases to unstructured text.

The primary focus of this internship is the conception and implementation of a scalable, flexible architecture. This architecture aims to bridge an LLM-based front-end with various back-end services to perform actionable tasks. Imagine an application where a user interacts with an LLM to query unstructured employee information, such as qualifications or project experience, stored in textual formats like PDFs or specialized CV tools. The system you will design is tasked with interpreting user queries and delivering relevant, understandable information.

The architectural framework you're tasked with conceiving should be adaptable enough to incorporate other technologies as they become relevant. To do so, you will investigate the existing options, compare them, and analyze what would be the best suited tools to use and orchestrate.

This internship offers a rich ground for hands-on experience, particularly in a research area aiming to expand the operational scope of conversational AI through thoughtful architectural design. This is both an educational and professional development opportunity in a rapidly evolving technological landscape.

Objectives: 

• To explore and implement architectures for integrating Large Language Models (like GPT) with actionable back-ends.
• To enable real-time, two-way interactions between users and databases/APIs through a chat interface powered by this architecture.
• To deliver a proof-of-concept application capable of conducting employee searches based on criteria such as skillsets, location, and availability.

• A dynamic work and collaborative environment with a highly motivated multi-cultural and international sites team
• Personal development through training and coaching
• A culture of collaboration across all disciplines
• The chance to make a difference in peoples’ life by building innovative solutions
• High innovation and research backed up by collaboration with universities like EPFL
• Various internal coding events (Hackathon, Brownbags), see our technical blog
• Monthly After-Works organized per locations
• Good work-life balance (2 days per week from home)

• Knowledge of software architecture design principles and patterns, with an emphasis on scalability and adaptability
• Familiarity with Natural Language Processing (NLP) or Machine Learning (ML), knowledge about LLMs
• Experience with Python (Pandas, FastAPI, Scikit-Learn…) and software engineering principles
• Understanding of RESTful APIs and database management systems (SQL or NoSQL)
• Strong communication skills in both French and English, capable of articulating complex ideas clearly in both written and verbal forms