Host Institution: University Paris-Est Créteil, LISSI laboratory
Industry Partner: Declic Startup – an innovative social networking and event organization platform
Supervision: Dr. Alice Othmani, HDR, Professor at University Paris-Est Créteil (UPEC)
Duration: 3 years
Location: Paris, France with potential collaborations in the Middle East and Europe
🔍 Context and Motivation
The rapid growth of social networking applications demands intelligent, personalized, and multilingual user experiences. Declic is a fast-growing platform designed to help individuals discover, organize, and promote social events and build meaningful real-world connections.
To address the challenges of personalization, multilingual interaction, user safety, and scalability, Declic seeks to integrate Large Language Models (LLMs) as a core technological component. This PhD project will focus on researching, developing, and deploying LLM-based solutions to enhance the user experience, automate content generation and moderation, and foster inclusive, personalized engagement on the platform.
This project offers a unique opportunity to combine cutting-edge AI research with real-world application and impact within a startup ecosystem.
🎯 Research Objectives
The PhD candidate will work on the following key objectives:
- Design and develop an LLM-powered personalized recommendation system for events, communities, and connections, leveraging user-generated data (profiles, preferences, activities, natural language descriptions).
- Build a multilingual conversational agent to assist users in discovering events, creating groups, and navigating the app through natural language queries (in Arabic, French, English).
- Implement intelligent, automated moderation systems using LLMs to detect inappropriate, harmful, or unsafe content across multiple languages.
- Investigate bias, fairness, and transparency in LLM-based recommendations and moderation within diverse, multilingual user populations.
- Explore automatic content generation for event descriptions, titles, and marketing texts, enabling event organizers to easily publish attractive and engaging content.
- Validate models in a production environment using Declic’s user interaction data and contribute to live deployments impacting 1M+ users.
🛠️ Key Research Challenges
- Fine-tuning and adapting large language models on sparse, noisy, user-generated social data
- Combining structured data (user profiles, event metadata) with unstructured text (bios, descriptions, chats)
- Achieving scalable inference on mobile and web platforms
- Addressing multilingual NLP challenges in Arabic, French, and English
- Balancing personalization and privacy, bias mitigation, and ethical AI in a social context
👩💻 Candidate Profile
We are seeking an outstanding, highly motivated candidate with:
✅ A Master’s degree in Artificial Intelligence, Natural Language Processing, Machine Learning, or related fields
✅ Strong programming skills in Python and experience with deep learning frameworks (PyTorch, TensorFlow)
✅ Knowledge of NLP and LLM architectures (Transformers, GPT, BERT, etc.)
✅ Interest in applied AI for social networking, recommender systems, conversational agents, or content moderation
✅ Excellent written and verbal communication skills (English required; French/Arabic a plus)
✅ Ability to work across academic and industry environments, collaborating with software engineers and business stakeholders
📚 Expected Outcomes
- Scientific contributions: High-impact publications in top AI/NLP conferences and journals
- Practical outcomes: Deployment of AI solutions in Declic’s production platform, contributing to user growth, engagement, and content safety
- Transferable innovations for broader applications in social networking, event management, and digital communities
💰 Funding and Benefits
- Fully funded PhD scholarship (tuition fees + stipend)
- Access to real-world production data and deployment pipelines
- International collaboration opportunities in Europe and the Middle East
- Support for conference participation and publications
🤝 Why this PhD?
This PhD offers a unique interdisciplinary research experience at the intersection of AI innovation, human-centered design, and social technologies, bridging academic excellence with real-world startup impact. The candidate will play a key role in shaping the future of intelligent, multilingual, inclusive social networking in Declic and beyond.
👉 Interested applicants should submit:
- A CV
- A motivation letter
- Academic transcripts
- Contact details of 2 referees
Send applications to: recrutement@declic.net
Application deadline: 30/06/2025
